Merge pull request #845 from Sh1n3zZ/gemini-embedding

feat: gemini Embeddings support

.env.example

@@ -10,9 +10,9 @@
 
 # 数据库相关配置
 # 数据库连接字符串
-# SQL_DSN=mysql://user:password@tcp(127.0.0.1:3306)/dbname?parseTime=true
+# SQL_DSN=user:password@tcp(127.0.0.1:3306)/dbname?parseTime=true
 # 日志数据库连接字符串
-# LOG_SQL_DSN=mysql://user:password@tcp(127.0.0.1:3306)/logdb?parseTime=true
+# LOG_SQL_DSN=user:password@tcp(127.0.0.1:3306)/logdb?parseTime=true
 # SQLite数据库路径
 # SQLITE_PATH=/path/to/sqlite.db
 # 数据库最大空闲连接数
@@ -50,10 +50,6 @@
 # CHANNEL_TEST_FREQUENCY=10
 # 生成默认token
 # GENERATE_DEFAULT_TOKEN=false
-# Gemini 安全设置
-# GEMINI_SAFETY_SETTING=BLOCK_NONE
-# Gemini版本设置
-# GEMINI_MODEL_MAP=gemini-1.0-pro:v1
 # Cohere 安全设置
 # COHERE_SAFETY_SETTING=NONE
 # 是否统计图片token

.github/workflows/docker-image-arm64.yml

@@ -13,7 +13,7 @@ on:
 jobs:
   push_to_registries:
     name: Push Docker image to multiple registries
-    runs-on: self-hosted
+    runs-on: ubuntu-latest
     permissions:
       packages: write
       contents: read

.github/workflows/macos-release.yml

@@ -9,7 +9,7 @@ on:
       - '!*-alpha*'
 jobs:
   release:
-    runs-on: ubuntu-latest
+    runs-on: macos-latest
     steps:
       - name: Checkout
         uses: actions/checkout@v3

.github/workflows/windows-release.yml

@@ -9,7 +9,7 @@ on:
       - '!*-alpha*'
 jobs:
   release:
-    runs-on: ubuntu-latest
+    runs-on: windows-latest
     defaults:
       run:
         shell: bash

Dockerfile

@@ -1,4 +1,4 @@
-FROM oven/bun:latest as builder
+FROM oven/bun:latest AS builder
 
 WORKDIR /build
 COPY web/package.json .
@@ -7,18 +7,20 @@ COPY ./web .
 COPY ./VERSION .
 RUN DISABLE_ESLINT_PLUGIN='true' VITE_REACT_APP_VERSION=$(cat VERSION) bun run build
 
-FROM golang AS builder2
+FROM golang:alpine AS builder2
 
 ENV GO111MODULE=on \
-    CGO_ENABLED=1 \
+    CGO_ENABLED=0 \
     GOOS=linux
 
 WORKDIR /build
+
 ADD go.mod go.sum ./
 RUN go mod download
+
 COPY . .
 COPY --from=builder /build/dist ./web/dist
-RUN go build -ldflags "-s -w -X 'one-api/common.Version=$(cat VERSION)' -extldflags '-static'" -o one-api
+RUN go build -ldflags "-s -w -X 'one-api/common.Version=$(cat VERSION)'" -o one-api
 
 FROM alpine
 

README.en.md

@@ -59,10 +59,24 @@
 13. 🎵 Added [Suno API](https://github.com/Suno-API/Suno-API) interface support, [Integration Guide](Suno.md)
 14. 🔄 Support for Rerank models, compatible with Cohere and Jina, can integrate with Dify, [Integration Guide](Rerank.md)
 15. ⚡ **[OpenAI Realtime API](https://platform.openai.com/docs/guides/realtime/integration)** - Support for OpenAI's Realtime API, including Azure channels
+16. 🧠 Support for setting reasoning effort through model name suffix:
+    - Add suffix `-high` to set high reasoning effort (e.g., `o3-mini-high`)
+    - Add suffix `-medium` to set medium reasoning effort
+    - Add suffix `-low` to set low reasoning effort
+17. 🔄 Thinking to content option `thinking_to_content` in `Channel->Edit->Channel Extra Settings`, default is `false`, when `true`, the `reasoning_content` of the thinking content will be converted to `<think>` tags and concatenated to the content returned.
+18. 🔄 Model rate limit, support setting total request limit and successful request limit in `System Settings->Rate Limit Settings`
+19. 💰 Cache billing support, when enabled can charge a configurable ratio for cache hits:
+    1. Set `Prompt Cache Ratio` in `System Settings -> Operation Settings`
+    2. Set `Prompt Cache Ratio` in channel settings, range 0-1 (e.g., 0.5 means 50% charge on cache hits)
+    3. Supported channels:
+        - [x] OpenAI
+        - [x] Azure 
+        - [x] DeepSeek
+        - [ ] Claude
 
 ## Model Support
 This version additionally supports:
-1. Third-party model **gps** (gpt-4-gizmo-*)
+1. Third-party model **gpts** (gpt-4-gizmo-*)
 2. [Midjourney-Proxy(Plus)](https://github.com/novicezk/midjourney-proxy) interface, [Integration Guide](Midjourney.md)
 3. Custom channels with full API URL support
 4. [Suno API](https://github.com/Suno-API/Suno-API) interface, [Integration Guide](Suno.md)
@@ -84,15 +98,15 @@ You can add custom models gpt-4-gizmo-* in channels. These are third-party model
 - `GEMINI_VISION_MAX_IMAGE_NUM`: Gemini model maximum image number, default `16`, set to `-1` to disable
 - `MAX_FILE_DOWNLOAD_MB`: Maximum file download size in MB, default `20`
 - `CRYPTO_SECRET`: Encryption key for encrypting database content
+- `AZURE_DEFAULT_API_VERSION`: Azure channel default API version, if not specified in channel settings, use this version, default `2024-12-01-preview`
+- `NOTIFICATION_LIMIT_DURATION_MINUTE`: Duration of notification limit in minutes, default `10`
+- `NOTIFY_LIMIT_COUNT`: Maximum number of user notifications in the specified duration, default `2`
 
 ## Deployment
+
 > [!TIP]
 > Latest Docker image: `calciumion/new-api:latest`  
-> Default account: root, password: 123456  
-> Update command:
-> ```
-> docker run --rm -v /var/run/docker.sock:/var/run/docker.sock containrrr/watchtower -cR
-> ```
+> Default account: root, password: 123456
 
 ### Multi-Server Deployment
 - Must set `SESSION_SECRET` environment variable, otherwise login state will not be consistent across multiple servers.
@@ -102,29 +116,61 @@ You can add custom models gpt-4-gizmo-* in channels. These are third-party model
 - Local database (default): SQLite (Docker deployment must mount `/data` directory)
 - Remote database: MySQL >= 5.7.8, PgSQL >= 9.6
 
+### Deployment with BT Panel
+Install BT Panel (**version 9.2.0** or above) from [BT Panel Official Website](https://www.bt.cn/new/download.html), choose the stable version script to download and install.  
+After installation, log in to BT Panel and click Docker in the menu bar. First-time access will prompt to install Docker service. Click Install Now and follow the prompts to complete installation.  
+After installation, find **New-API** in the app store, click install, configure basic options to complete installation.  
+[Pictorial Guide](BT.md)
+
 ### Docker Deployment
+
 ### Using Docker Compose (Recommended)
 ```shell
 # Clone project
 git clone https://github.com/Calcium-Ion/new-api.git
 cd new-api
 # Edit docker-compose.yml as needed
+# nano docker-compose.yml
+# vim docker-compose.yml
 # Start
 docker-compose up -d
 ```
 
+#### Update Version
+```shell
+docker-compose pull
+docker-compose up -d
+```
+
 ### Direct Docker Image Usage
 ```shell
 # SQLite deployment:
 docker run --name new-api -d --restart always -p 3000:3000 -e TZ=Asia/Shanghai -v /home/ubuntu/data/new-api:/data calciumion/new-api:latest
+
 # MySQL deployment (add -e SQL_DSN="root:123456@tcp(localhost:3306)/oneapi"), modify database connection parameters as needed
 # Example:
 docker run --name new-api -d --restart always -p 3000:3000 -e SQL_DSN="root:123456@tcp(localhost:3306)/oneapi" -e TZ=Asia/Shanghai -v /home/ubuntu/data/new-api:/data calciumion/new-api:latest
 ```
 
+#### Update Version
+```shell
+# Pull the latest image
+docker pull calciumion/new-api:latest
+# Stop and remove the old container
+docker stop new-api
+docker rm new-api
+# Run the new container with the same parameters as before
+docker run --name new-api -d --restart always -p 3000:3000 -e TZ=Asia/Shanghai -v /home/ubuntu/data/new-api:/data calciumion/new-api:latest
+```
+
+Alternatively, you can use Watchtower for automatic updates (not recommended, may cause database incompatibility):
+```shell
+docker run --rm -v /var/run/docker.sock:/var/run/docker.sock containrrr/watchtower -cR
+```
+
 ## Channel Retry
 Channel retry is implemented, configurable in `Settings->Operation Settings->General Settings`. **Cache recommended**.  
-First retry uses same priority, second retry uses next priority, and so on.
+If retry is enabled, the system will automatically use the next priority channel for the same request after a failed request.
 
 ### Cache Configuration
 1. `REDIS_CONN_STRING`: Use Redis as cache

README.md

@@ -65,10 +65,27 @@
 14. 🔄 支持Rerank模型，目前兼容Cohere和Jina，可接入Dify，[对接文档](Rerank.md)
 15. ⚡ **[OpenAI Realtime API](https://platform.openai.com/docs/guides/realtime/integration)** - 支持OpenAI的Realtime API，支持Azure渠道
 16. 支持使用路由/chat2link 进入聊天界面
+17. 🧠 支持通过模型名称后缀设置 reasoning effort：
+    1. OpenAI o系列模型
+        - 添加后缀 `-high` 设置为 high reasoning effort (例如: `o3-mini-high`)
+        - 添加后缀 `-medium` 设置为 medium reasoning effort (例如: `o3-mini-medium`)
+        - 添加后缀 `-low` 设置为 low reasoning effort (例如: `o3-mini-low`)
+    2. Claude 思考模型
+        - 添加后缀 `-thinking` 启用思考模式 (例如: `claude-3-7-sonnet-20250219-thinking`)
+18. 🔄 思考转内容，支持在 `渠道-编辑-渠道额外设置` 中设置 `thinking_to_content` 选项，默认`false`，开启后会将思考内容`reasoning_content`转换为`<think>`标签拼接到内容中返回。
+19. 🔄 模型限流，支持在 `系统设置-速率限制设置` 中设置模型限流，支持设置总请求数限制和成功请求数限制
+20. 💰 缓存计费支持，开启后可以在缓存命中时按照设定的比例计费：
+    1. 在 `系统设置-运营设置` 中设置 `提示缓存倍率` 选项
+    2. 在渠道中设置 `提示缓存倍率`，范围 0-1，例如设置为 0.5 表示缓存命中时按照 50% 计费
+    3. 支持的渠道：
+        - [x] OpenAI
+        - [x] Azure
+        - [x] DeepSeek
+        - [ ] Claude
 
 ## 模型支持
 此版本额外支持以下模型：
-1. 第三方模型 **gps** （gpt-4-gizmo-*）
+1. 第三方模型 **gpts** （gpt-4-gizmo-*）
 2. [Midjourney-Proxy(Plus)](https://github.com/novicezk/midjourney-proxy)接口，[对接文档](Midjourney.md)
 3. 自定义渠道，支持填入完整调用地址
 4. [Suno API](https://github.com/Suno-API/Suno-API) 接口，[对接文档](Suno.md)
@@ -85,19 +102,23 @@
 - `GET_MEDIA_TOKEN`：是否统计图片token，默认为 `true`，关闭后将不再在本地计算图片token，可能会导致和上游计费不同，此项覆盖 `GET_MEDIA_TOKEN_NOT_STREAM` 选项作用。
 - `GET_MEDIA_TOKEN_NOT_STREAM`：是否在非流（`stream=false`）情况下统计图片token，默认为 `true`。
 - `UPDATE_TASK`：是否更新异步任务（Midjourney、Suno），默认为 `true`，关闭后将不会更新任务进度。
-- `GEMINI_MODEL_MAP`：Gemini模型指定版本(v1/v1beta)，使用“模型:版本”指定，","分隔，例如：-e GEMINI_MODEL_MAP="gemini-1.5-pro-latest:v1beta,gemini-1.5-pro-001:v1beta"，为空则使用默认配置(v1beta)
-- `COHERE_SAFETY_SETTING`：Cohere模型[安全设置](https://docs.cohere.com/docs/safety-modes#overview)，可选值为 `NONE`, `CONTEXTUAL`，`STRICT`，默认为 `NONE`。
+- `COHERE_SAFETY_SETTING`：Cohere模型[安全设置](https://docs.cohere.com/docs/safety-modes#overview)，可选值为 `NONE`, `CONTEXTUAL`, `STRICT`，默认为 `NONE`。
 - `GEMINI_VISION_MAX_IMAGE_NUM`：Gemini模型最大图片数量，默认为 `16`，设置为 `-1` 则不限制。
 - `MAX_FILE_DOWNLOAD_MB`: 最大文件下载大小，单位 MB，默认为 `20`。
 - `CRYPTO_SECRET`：加密密钥，用于加密数据库内容。
+- `AZURE_DEFAULT_API_VERSION`：Azure渠道默认API版本，如果渠道设置中未指定API版本，则使用此版本，默认为 `2024-12-01-preview`
+- `NOTIFICATION_LIMIT_DURATION_MINUTE`：通知限制的持续时间（分钟），默认为 `10`。
+- `NOTIFY_LIMIT_COUNT`：用户通知在指定持续时间内的最大数量，默认为 `2`。
+
+## 已废弃的环境变量
+- ~~`GEMINI_MODEL_MAP`（已废弃）~~：改为到`设置-模型相关设置`中设置
+- ~~`GEMINI_SAFETY_SETTING`（已废弃）~~：改为到`设置-模型相关设置`中设置
+
 ## 部署
+
 > [!TIP]
 > 最新版Docker镜像：`calciumion/new-api:latest`  
-> 默认账号root 密码123456  
-> 更新指令：
-> ```
-> docker run --rm -v /var/run/docker.sock:/var/run/docker.sock containrrr/watchtower -cR
-> ```
+> 默认账号root 密码123456
 
 ### 多机部署
 - 必须设置环境变量 `SESSION_SECRET`，否则会导致多机部署时登录状态不一致。
@@ -114,28 +135,57 @@
 [图文教程](BT.md)
 
 ### 基于 Docker 进行部署
+
+> [!TIP]
+> 默认管理员账号root 密码123456
+
 ### 使用 Docker Compose 部署（推荐）
 ```shell
 # 下载项目
 git clone https://github.com/Calcium-Ion/new-api.git
 cd new-api
 # 按需编辑 docker-compose.yml
+# nano docker-compose.yml
+# vim docker-compose.yml
 # 启动
 docker-compose up -d
 ```
 
+#### 更新版本
+```shell
+docker-compose pull
+docker-compose up -d
+```
+
 ### 直接使用 Docker 镜像
 ```shell
 # 使用 SQLite 的部署命令：
 docker run --name new-api -d --restart always -p 3000:3000 -e TZ=Asia/Shanghai -v /home/ubuntu/data/new-api:/data calciumion/new-api:latest
+
 # 使用 MySQL 的部署命令，在上面的基础上添加 `-e SQL_DSN="root:123456@tcp(localhost:3306)/oneapi"`，请自行修改数据库连接参数。
 # 例如：
 docker run --name new-api -d --restart always -p 3000:3000 -e SQL_DSN="root:123456@tcp(localhost:3306)/oneapi" -e TZ=Asia/Shanghai -v /home/ubuntu/data/new-api:/data calciumion/new-api:latest
 ```
 
+#### 更新版本
+```shell
+# 拉取最新镜像
+docker pull calciumion/new-api:latest
+# 停止并删除旧容器
+docker stop new-api
+docker rm new-api
+# 使用相同参数运行新容器
+docker run --name new-api -d --restart always -p 3000:3000 -e TZ=Asia/Shanghai -v /home/ubuntu/data/new-api:/data calciumion/new-api:latest
+```
+
+或者使用 Watchtower 自动更新（不推荐，可能会导致数据库不兼容）：
+```shell
+docker run --rm -v /var/run/docker.sock:/var/run/docker.sock containrrr/watchtower -cR
+```
+
 ## 渠道重试
 渠道重试功能已经实现，可以在`设置->运营设置->通用设置`设置重试次数，**建议开启缓存**功能。  
-如果开启了重试功能，第一次重试使用同优先级，第二次重试使用下一个优先级，以此类推。
+如果开启了重试功能，重试使用下一个优先级，以此类推。
 ### 缓存设置方法
 1. `REDIS_CONN_STRING`：设置之后将使用 Redis 作为缓存使用。
     + 例子：`REDIS_CONN_STRING=redis://default:redispw@localhost:49153`
@@ -178,8 +228,8 @@ docker run --name new-api -d --restart always -p 3000:3000 -e SQL_DSN="root:1234
 - [neko-api-key-tool](https://github.com/Calcium-Ion/neko-api-key-tool)：用key查询使用额度
 
 其他基于New API的项目：
-- [new-api-horizon](https://github.com/Calcium-Ion/new-api-horizon)：New API高性能优化版，并支持Claude格式
-- [VoAPI](https://github.com/VoAPI/VoAPI)：基于New API的闭源项目
+- [new-api-horizon](https://github.com/Calcium-Ion/new-api-horizon)：New API高性能优化版，专注于高并发优化，并支持Claude格式
+- [VoAPI](https://github.com/VoAPI/VoAPI)：基于New API的前端美化版本，闭源免费
 
 ## 🌟 Star History
 

Rerank.md

@@ -13,7 +13,7 @@ Request:
 
 ```json
 {
-  "model": "rerank-multilingual-v3.0",
+  "model": "jina-reranker-v2-base-multilingual",
   "query": "What is the capital of the United States?",
   "top_n": 3,
   "documents": [

common/constants.go

@@ -15,8 +15,9 @@ var SystemName = "New API"
 var Footer = ""
 var Logo = ""
 var TopUpLink = ""
-var ChatLink = ""
-var ChatLink2 = ""
+
+// var ChatLink = ""
+// var ChatLink2 = ""
 var QuotaPerUnit = 500 * 1000.0 // $0.002 / 1K tokens
 var DisplayInCurrencyEnabled = true
 var DisplayTokenStatEnabled = true
@@ -101,7 +102,7 @@ var PreConsumedQuota = 500
 
 var RetryTimes = 0
 
-var RootUserEmail = ""
+//var RootUserEmail = ""
 
 var IsMasterNode = os.Getenv("NODE_TYPE") != "slave"
 
@@ -231,8 +232,10 @@ const (
 	ChannelTypeVertexAi       = 41
 	ChannelTypeMistral        = 42
 	ChannelTypeDeepSeek       = 43
-
-	ChannelTypeDummy // this one is only for count, do not add any channel after this
+	ChannelTypeMokaAI         = 44
+	ChannelTypeVolcEngine     = 45
+	ChannelTypeBaiduV2        = 46
+	ChannelTypeDummy          // this one is only for count, do not add any channel after this
 
 )
 
@@ -274,11 +277,14 @@ var ChannelBaseURLs = []string{
 	"https://api.cohere.ai",                     //34
 	"https://api.minimax.chat",                  //35
 	"",                                          //36
-	"",                                          //37
+	"https://api.dify.ai",                       //37
 	"https://api.jina.ai",                       //38
 	"https://api.cloudflare.com",                //39
 	"https://api.siliconflow.cn",                //40
 	"",                                          //41
 	"https://api.mistral.ai",                    //42
 	"https://api.deepseek.com",                  //43
+	"https://api.moka.ai",                       //44
+	"https://ark.cn-beijing.volces.com",         //45
+	"https://qianfan.baidubce.com",              //46
 }

common/database.go

@@ -3,5 +3,6 @@ package common
 var UsingSQLite = false
 var UsingPostgreSQL = false
 var UsingMySQL = false
+var UsingClickHouse = false
 
 var SQLitePath = "one-api.db?_busy_timeout=5000"

common/go-channel.go

@@ -1,22 +1,9 @@
 package common
 
 import (
-	"fmt"
-	"runtime/debug"
 	"time"
 )
 
-func SafeGoroutine(f func()) {
-	go func() {
-		defer func() {
-			if r := recover(); r != nil {
-				SysError(fmt.Sprintf("child goroutine panic occured: error: %v, stack: %s", r, string(debug.Stack())))
-			}
-		}()
-		f()
-	}()
-}
-
 func SafeSendBool(ch chan bool, value bool) (closed bool) {
 	defer func() {
 		// Recover from panic if one occured. A panic would mean the channel was closed.

common/gopool.go

@@ -0,0 +1,24 @@
+package common
+
+import (
+	"context"
+	"fmt"
+	"github.com/bytedance/gopkg/util/gopool"
+	"math"
+)
+
+var relayGoPool gopool.Pool
+
+func init() {
+	relayGoPool = gopool.NewPool("gopool.RelayPool", math.MaxInt32, gopool.NewConfig())
+	relayGoPool.SetPanicHandler(func(ctx context.Context, i interface{}) {
+		if stopChan, ok := ctx.Value("stop_chan").(chan bool); ok {
+			SafeSendBool(stopChan, true)
+		}
+		SysError(fmt.Sprintf("panic in gopool.RelayPool: %v", i))
+	})
+}
+
+func RelayCtxGo(ctx context.Context, f func()) {
+	relayGoPool.CtxGo(ctx, f)
+}

common/logger.go

@@ -4,6 +4,7 @@ import (
 	"context"
 	"encoding/json"
 	"fmt"
+	"github.com/bytedance/gopkg/util/gopool"
 	"github.com/gin-gonic/gin"
 	"io"
 	"log"
@@ -80,9 +81,9 @@ func logHelper(ctx context.Context, level string, msg string) {
 	if logCount > maxLogCount && !setupLogWorking {
 		logCount = 0
 		setupLogWorking = true
-		go func() {
+		gopool.Go(func() {
 			SetupLogger()
-		}()
+		})
 	}
 }
 
@@ -100,6 +101,14 @@ func LogQuota(quota int) string {
 	}
 }
 
+func FormatQuota(quota int) string {
+	if DisplayInCurrencyEnabled {
+		return fmt.Sprintf("＄%.6f", float64(quota)/QuotaPerUnit)
+	} else {
+		return fmt.Sprintf("%d", quota)
+	}
+}
+
 // LogJson 仅供测试使用 only for test
 func LogJson(ctx context.Context, msg string, obj any) {
 	jsonStr, err := json.Marshal(obj)

common/redis.go

@@ -32,6 +32,7 @@ func InitRedisClient() (err error) {
 	if err != nil {
 		FatalLog("failed to parse Redis connection string: " + err.Error())
 	}
+	opt.PoolSize = GetEnvOrDefault("REDIS_POOL_SIZE", 10)
 	RDB = redis.NewClient(opt)
 
 	ctx, cancel := context.WithTimeout(context.Background(), 5*time.Second)
@@ -41,6 +42,10 @@ func InitRedisClient() (err error) {
 	if err != nil {
 		FatalLog("Redis ping test failed: " + err.Error())
 	}
+	if DebugEnabled {
+		SysLog(fmt.Sprintf("Redis connected to %s", opt.Addr))
+		SysLog(fmt.Sprintf("Redis database: %d", opt.DB))
+	}
 	return err
 }
 
@@ -53,13 +58,20 @@ func ParseRedisOption() *redis.Options {
 }
 
 func RedisSet(key string, value string, expiration time.Duration) error {
+	if DebugEnabled {
+		SysLog(fmt.Sprintf("Redis SET: key=%s, value=%s, expiration=%v", key, value, expiration))
+	}
 	ctx := context.Background()
 	return RDB.Set(ctx, key, value, expiration).Err()
 }
 
 func RedisGet(key string) (string, error) {
+	if DebugEnabled {
+		SysLog(fmt.Sprintf("Redis GET: key=%s", key))
+	}
 	ctx := context.Background()
-	return RDB.Get(ctx, key).Result()
+	val, err := RDB.Get(ctx, key).Result()
+	return val, err
 }
 
 //func RedisExpire(key string, expiration time.Duration) error {
@@ -73,16 +85,25 @@ func RedisGet(key string) (string, error) {
 //}
 
 func RedisDel(key string) error {
+	if DebugEnabled {
+		SysLog(fmt.Sprintf("Redis DEL: key=%s", key))
+	}
 	ctx := context.Background()
 	return RDB.Del(ctx, key).Err()
 }
 
 func RedisHDelObj(key string) error {
+	if DebugEnabled {
+		SysLog(fmt.Sprintf("Redis HDEL: key=%s", key))
+	}
 	ctx := context.Background()
 	return RDB.HDel(ctx, key).Err()
 }
 
 func RedisHSetObj(key string, obj interface{}, expiration time.Duration) error {
+	if DebugEnabled {
+		SysLog(fmt.Sprintf("Redis HSET: key=%s, obj=%+v, expiration=%v", key, obj, expiration))
+	}
 	ctx := context.Background()
 
 	data := make(map[string]interface{})
@@ -130,6 +151,9 @@ func RedisHSetObj(key string, obj interface{}, expiration time.Duration) error {
 }
 
 func RedisHGetObj(key string, obj interface{}) error {
+	if DebugEnabled {
+		SysLog(fmt.Sprintf("Redis HGETALL: key=%s", key))
+	}
 	ctx := context.Background()
 
 	result, err := RDB.HGetAll(ctx, key).Result()
@@ -208,6 +232,9 @@ func RedisHGetObj(key string, obj interface{}) error {
 
 // RedisIncr Add this function to handle atomic increments
 func RedisIncr(key string, delta int64) error {
+	if DebugEnabled {
+		SysLog(fmt.Sprintf("Redis INCR: key=%s, delta=%d", key, delta))
+	}
 	// 检查键的剩余生存时间
 	ttlCmd := RDB.TTL(context.Background(), key)
 	ttl, err := ttlCmd.Result()
@@ -238,6 +265,9 @@ func RedisIncr(key string, delta int64) error {
 }
 
 func RedisHIncrBy(key, field string, delta int64) error {
+	if DebugEnabled {
+		SysLog(fmt.Sprintf("Redis HINCRBY: key=%s, field=%s, delta=%d", key, field, delta))
+	}
 	ttlCmd := RDB.TTL(context.Background(), key)
 	ttl, err := ttlCmd.Result()
 	if err != nil && !errors.Is(err, redis.Nil) {
@@ -262,6 +292,9 @@ func RedisHIncrBy(key, field string, delta int64) error {
 }
 
 func RedisHSetField(key, field string, value interface{}) error {
+	if DebugEnabled {
+		SysLog(fmt.Sprintf("Redis HSET field: key=%s, field=%s, value=%v", key, field, value))
+	}
 	ttlCmd := RDB.TTL(context.Background(), key)
 	ttl, err := ttlCmd.Result()
 	if err != nil && !errors.Is(err, redis.Nil) {

common/utils.go

@@ -5,6 +5,7 @@ import (
 	"context"
 	crand "crypto/rand"
 	"encoding/base64"
+	"encoding/json"
 	"fmt"
 	"github.com/pkg/errors"
 	"html/template"
@@ -213,6 +214,24 @@ func RandomSleep() {
 	time.Sleep(time.Duration(rand.Intn(3000)) * time.Millisecond)
 }
 
+func GetPointer[T any](v T) *T {
+	return &v
+}
+
+func Any2Type[T any](data any) (T, error) {
+	var zero T
+	bytes, err := json.Marshal(data)
+	if err != nil {
+		return zero, err
+	}
+	var res T
+	err = json.Unmarshal(bytes, &res)
+	if err != nil {
+		return zero, err
+	}
+	return res, nil
+}
+
 // SaveTmpFile saves data to a temporary file. The filename would be apppended with a random string.
 func SaveTmpFile(filename string, data io.Reader) (string, error) {
 	f, err := os.CreateTemp(os.TempDir(), filename)

constant/channel_setting.go

@@ -1,5 +1,7 @@
 package constant
 
 var (
-	ForceFormat = "force_format" // ForceFormat 强制格式化为OpenAI格式
+	ForceFormat                     = "force_format"        // ForceFormat 强制格式化为OpenAI格式
+	ChanelSettingProxy              = "proxy"               // Proxy 代理
+	ChannelSettingThinkingToContent = "thinking_to_content" // ThinkingToContent
 )

constant/context_key.go

@@ -2,4 +2,9 @@ package constant
 
 const (
 	ContextKeyRequestStartTime = "request_start_time"
+	ContextKeyUserSetting      = "user_setting"
+	ContextKeyUserQuota        = "user_quota"
+	ContextKeyUserStatus       = "user_status"
+	ContextKeyUserEmail        = "user_email"
+	ContextKeyUserGroup        = "user_group"
 )

constant/env.go

@@ -1,10 +1,7 @@
 package constant
 
 import (
-	"fmt"
 	"one-api/common"
-	"os"
-	"strings"
 )
 
 var StreamingTimeout = common.GetEnvOrDefault("STREAMING_TIMEOUT", 60)
@@ -21,26 +18,31 @@ var GetMediaTokenNotStream = common.GetEnvOrDefaultBool("GET_MEDIA_TOKEN_NOT_STR
 
 var UpdateTask = common.GetEnvOrDefaultBool("UPDATE_TASK", true)
 
-var GeminiModelMap = map[string]string{
-	"gemini-1.0-pro": "v1",
-}
+var AzureDefaultAPIVersion = common.GetEnvOrDefaultString("AZURE_DEFAULT_API_VERSION", "2024-12-01-preview")
+
+//var GeminiModelMap = map[string]string{
+//	"gemini-1.0-pro": "v1",
+//}
 
 var GeminiVisionMaxImageNum = common.GetEnvOrDefault("GEMINI_VISION_MAX_IMAGE_NUM", 16)
 
+var NotifyLimitCount = common.GetEnvOrDefault("NOTIFY_LIMIT_COUNT", 2)
+var NotificationLimitDurationMinute = common.GetEnvOrDefault("NOTIFICATION_LIMIT_DURATION_MINUTE", 10)
+
 func InitEnv() {
-	modelVersionMapStr := strings.TrimSpace(os.Getenv("GEMINI_MODEL_MAP"))
-	if modelVersionMapStr == "" {
-		return
-	}
-	for _, pair := range strings.Split(modelVersionMapStr, ",") {
-		parts := strings.Split(pair, ":")
-		if len(parts) == 2 {
-			GeminiModelMap[parts[0]] = parts[1]
-		} else {
-			common.SysError(fmt.Sprintf("invalid model version map: %s", pair))
-		}
-	}
+	//modelVersionMapStr := strings.TrimSpace(os.Getenv("GEMINI_MODEL_MAP"))
+	//if modelVersionMapStr == "" {
+	//	return
+	//}
+	//for _, pair := range strings.Split(modelVersionMapStr, ",") {
+	//	parts := strings.Split(pair, ":")
+	//	if len(parts) == 2 {
+	//		GeminiModelMap[parts[0]] = parts[1]
+	//	} else {
+	//		common.SysError(fmt.Sprintf("invalid model version map: %s", pair))
+	//	}
+	//}
 }
 
-// 是否生成初始令牌，默认关闭。
+// GenerateDefaultToken 是否生成初始令牌，默认关闭。
 var GenerateDefaultToken = common.GetEnvOrDefaultBool("GENERATE_DEFAULT_TOKEN", false)

constant/user_setting.go

@@ -0,0 +1,14 @@
+package constant
+
+var (
+	UserSettingNotifyType            = "notify_type"             // QuotaWarningType 额度预警类型
+	UserSettingQuotaWarningThreshold = "quota_warning_threshold" // QuotaWarningThreshold 额度预警阈值
+	UserSettingWebhookUrl            = "webhook_url"             // WebhookUrl webhook地址
+	UserSettingWebhookSecret         = "webhook_secret"          // WebhookSecret webhook密钥
+	UserSettingNotificationEmail     = "notification_email"      // NotificationEmail 通知邮箱地址
+)
+
+var (
+	NotifyTypeEmail   = "email"   // Email 邮件
+	NotifyTypeWebhook = "webhook" // Webhook
+)

controller/channel-test.go

@@ -17,6 +17,7 @@ import (
 	"one-api/relay"
 	relaycommon "one-api/relay/common"
 	"one-api/relay/constant"
+	"one-api/relay/helper"
 	"one-api/service"
 	"strconv"
 	"strings"
@@ -41,9 +42,21 @@ func testChannel(channel *model.Channel, testModel string) (err error, openAIErr
 	}
 	w := httptest.NewRecorder()
 	c, _ := gin.CreateTestContext(w)
+
+	requestPath := "/v1/chat/completions"
+
+	// 先判断是否为 Embedding 模型
+	if strings.Contains(strings.ToLower(testModel), "embedding") ||
+		strings.HasPrefix(testModel, "m3e") || // m3e 系列模型
+		strings.Contains(testModel, "bge-") || // bge 系列模型
+		strings.Contains(testModel, "embed") ||
+		channel.Type == common.ChannelTypeMokaAI { // 其他 embedding 模型
+		requestPath = "/v1/embeddings" // 修改请求路径
+	}
+
 	c.Request = &http.Request{
 		Method: "POST",
-		URL:    &url.URL{Path: "/v1/chat/completions"},
+		URL:    &url.URL{Path: requestPath}, // 使用动态路径
 		Body:   nil,
 		Header: make(http.Header),
 	}
@@ -55,31 +68,34 @@ func testChannel(channel *model.Channel, testModel string) (err error, openAIErr
 			if len(channel.GetModels()) > 0 {
 				testModel = channel.GetModels()[0]
 			} else {
-				testModel = "gpt-3.5-turbo"
-			}
-		}
-	} else {
-		modelMapping := *channel.ModelMapping
-		if modelMapping != "" && modelMapping != "{}" {
-			modelMap := make(map[string]string)
-			err := json.Unmarshal([]byte(modelMapping), &modelMap)
-			if err != nil {
-				return err, service.OpenAIErrorWrapperLocal(err, "unmarshal_model_mapping_failed", http.StatusInternalServerError)
-			}
-			if modelMap[testModel] != "" {
-				testModel = modelMap[testModel]
+				testModel = "gpt-4o-mini"
 			}
 		}
 	}
 
+	cache, err := model.GetUserCache(1)
+	if err != nil {
+		return err, nil
+	}
+	cache.WriteContext(c)
+
 	c.Request.Header.Set("Authorization", "Bearer "+channel.Key)
 	c.Request.Header.Set("Content-Type", "application/json")
 	c.Set("channel", channel.Type)
 	c.Set("base_url", channel.GetBaseURL())
+	group, _ := model.GetUserGroup(1, false)
+	c.Set("group", group)
 
 	middleware.SetupContextForSelectedChannel(c, channel, testModel)
 
-	meta := relaycommon.GenRelayInfo(c)
+	info := relaycommon.GenRelayInfo(c)
+
+	err = helper.ModelMappedHelper(c, info)
+	if err != nil {
+		return err, nil
+	}
+	testModel = info.UpstreamModelName
+
 	apiType, _ := constant.ChannelType2APIType(channel.Type)
 	adaptor := relay.GetAdaptor(apiType)
 	if adaptor == nil {
@@ -87,12 +103,11 @@ func testChannel(channel *model.Channel, testModel string) (err error, openAIErr
 	}
 
 	request := buildTestRequest(testModel)
-	meta.UpstreamModelName = testModel
-	common.SysLog(fmt.Sprintf("testing channel %d with model %s", channel.Id, testModel))
+	common.SysLog(fmt.Sprintf("testing channel %d with model %s , info %v ", channel.Id, testModel, info))
 
-	adaptor.Init(meta)
+	adaptor.Init(info)
 
-	convertedRequest, err := adaptor.ConvertRequest(c, meta, request)
+	convertedRequest, err := adaptor.ConvertRequest(c, info, request)
 	if err != nil {
 		return err, nil
 	}
@@ -102,7 +117,7 @@ func testChannel(channel *model.Channel, testModel string) (err error, openAIErr
 	}
 	requestBody := bytes.NewBuffer(jsonData)
 	c.Request.Body = io.NopCloser(requestBody)
-	resp, err := adaptor.DoRequest(c, meta, requestBody)
+	resp, err := adaptor.DoRequest(c, info, requestBody)
 	if err != nil {
 		return err, nil
 	}
@@ -114,7 +129,7 @@ func testChannel(channel *model.Channel, testModel string) (err error, openAIErr
 			return fmt.Errorf("status code %d: %s", httpResp.StatusCode, err.Error.Message), err
 		}
 	}
-	usageA, respErr := adaptor.DoResponse(c, httpResp, meta)
+	usageA, respErr := adaptor.DoResponse(c, httpResp, info)
 	if respErr != nil {
 		return fmt.Errorf("%s", respErr.Error.Message), respErr
 	}
@@ -127,26 +142,28 @@ func testChannel(channel *model.Channel, testModel string) (err error, openAIErr
 	if err != nil {
 		return err, nil
 	}
-	modelPrice, usePrice := common.GetModelPrice(testModel, false)
-	modelRatio := common.GetModelRatio(testModel)
-	completionRatio := common.GetCompletionRatio(testModel)
-	ratio := modelRatio
+	info.PromptTokens = usage.PromptTokens
+	priceData, err := helper.ModelPriceHelper(c, info, usage.PromptTokens, int(request.MaxTokens))
+	if err != nil {
+		return err, nil
+	}
 	quota := 0
-	if !usePrice {
-		quota = usage.PromptTokens + int(math.Round(float64(usage.CompletionTokens)*completionRatio))
-		quota = int(math.Round(float64(quota) * ratio))
-		if ratio != 0 && quota <= 0 {
+	if !priceData.UsePrice {
+		quota = usage.PromptTokens + int(math.Round(float64(usage.CompletionTokens)*priceData.CompletionRatio))
+		quota = int(math.Round(float64(quota) * priceData.ModelRatio))
+		if priceData.ModelRatio != 0 && quota <= 0 {
 			quota = 1
 		}
 	} else {
-		quota = int(modelPrice * common.QuotaPerUnit)
+		quota = int(priceData.ModelPrice * common.QuotaPerUnit)
 	}
 	tok := time.Now()
 	milliseconds := tok.Sub(tik).Milliseconds()
 	consumedTime := float64(milliseconds) / 1000.0
-	other := service.GenerateTextOtherInfo(c, meta, modelRatio, 1, completionRatio, modelPrice)
-	model.RecordConsumeLog(c, 1, channel.Id, usage.PromptTokens, usage.CompletionTokens, testModel, "模型测试",
-		quota, "模型测试", 0, quota, int(consumedTime), false, "default", other)
+	other := service.GenerateTextOtherInfo(c, info, priceData.ModelRatio, priceData.GroupRatio, priceData.CompletionRatio,
+		usage.PromptTokensDetails.CachedTokens, priceData.CacheRatio, priceData.ModelPrice)
+	model.RecordConsumeLog(c, 1, channel.Id, usage.PromptTokens, usage.CompletionTokens, info.OriginModelName, "模型测试",
+		quota, "模型测试", 0, quota, int(consumedTime), false, info.Group, other)
 	common.SysLog(fmt.Sprintf("testing channel #%d, response: \n%s", channel.Id, string(respBody)))
 	return nil, nil
 }
@@ -156,12 +173,23 @@ func buildTestRequest(model string) *dto.GeneralOpenAIRequest {
 		Model:  "", // this will be set later
 		Stream: false,
 	}
-	if strings.HasPrefix(model, "o1") {
+
+	// 先判断是否为 Embedding 模型
+	if strings.Contains(strings.ToLower(model), "embedding") || // 其他 embedding 模型
+		strings.HasPrefix(model, "m3e") || // m3e 系列模型
+		strings.Contains(model, "bge-") {
+		testRequest.Model = model
+		// Embedding 请求
+		testRequest.Input = []string{"hello world"}
+		return testRequest
+	}
+	// 并非Embedding 模型
+	if strings.HasPrefix(model, "o1") || strings.HasPrefix(model, "o3") {
 		testRequest.MaxCompletionTokens = 10
-	} else if strings.HasPrefix(model, "gemini-2.0-flash-thinking") {
-		testRequest.MaxTokens = 2
+	} else if strings.Contains(model, "thinking") {
+		testRequest.MaxTokens = 50
 	} else {
-		testRequest.MaxTokens = 1
+		testRequest.MaxTokens = 10
 	}
 	content, _ := json.Marshal("hi")
 	testMessage := dto.Message{
@@ -217,9 +245,7 @@ var testAllChannelsLock sync.Mutex
 var testAllChannelsRunning bool = false
 
 func testAllChannels(notify bool) error {
-	if common.RootUserEmail == "" {
-		common.RootUserEmail = model.GetRootUserEmail()
-	}
+
 	testAllChannelsLock.Lock()
 	if testAllChannelsRunning {
 		testAllChannelsLock.Unlock()
@@ -274,10 +300,7 @@ func testAllChannels(notify bool) error {
 		testAllChannelsRunning = false
 		testAllChannelsLock.Unlock()
 		if notify {
-			err := common.SendEmail("通道测试完成", common.RootUserEmail, "通道测试完成，如果没有收到禁用通知，说明所有通道都正常")
-			if err != nil {
-				common.SysError(fmt.Sprintf("failed to send email: %s", err.Error()))
-			}
+			service.NotifyRootUser(dto.NotifyTypeChannelTest, "通道测试完成", "所有通道测试已完成")
 		}
 	})
 	return nil

controller/midjourney.go

@@ -159,7 +159,7 @@ func UpdateMidjourneyTaskBulk() {
 					common.LogError(ctx, "UpdateMidjourneyTask task error: "+err.Error())
 				} else {
 					if shouldReturnQuota {
-						err = model.IncreaseUserQuota(task.UserId, task.Quota)
+						err = model.IncreaseUserQuota(task.UserId, task.Quota, false)
 						if err != nil {
 							common.LogError(ctx, "fail to increase user quota: "+err.Error())
 						}

controller/misc.go

@@ -7,6 +7,7 @@ import (
 	"one-api/common"
 	"one-api/model"
 	"one-api/setting"
+	"one-api/setting/operation_setting"
 	"strings"
 
 	"github.com/gin-gonic/gin"
@@ -53,8 +54,7 @@ func GetStatus(c *gin.Context) {
 			"turnstile_check":          common.TurnstileCheckEnabled,
 			"turnstile_site_key":       common.TurnstileSiteKey,
 			"top_up_link":              common.TopUpLink,
-			"chat_link":                common.ChatLink,
-			"chat_link2":               common.ChatLink2,
+			"docs_link":                operation_setting.GetGeneralSetting().DocsLink,
 			"quota_per_unit":           common.QuotaPerUnit,
 			"display_in_currency":      common.DisplayInCurrencyEnabled,
 			"enable_batch_update":      common.BatchUpdateEnabled,
@@ -66,6 +66,8 @@ func GetStatus(c *gin.Context) {
 			"enable_online_topup":      setting.PayAddress != "" && setting.EpayId != "" && setting.EpayKey != "",
 			"mj_notify_enabled":        setting.MjNotifyEnabled,
 			"chats":                    setting.Chats,
+			"demo_site_enabled":        operation_setting.DemoSiteEnabled,
+			"self_use_mode_enabled":    operation_setting.SelfUseModeEnabled,
 		},
 	})
 	return

controller/model.go

@@ -216,6 +216,13 @@ func DashboardListModels(c *gin.Context) {
 	})
 }
 
+func EnabledListModels(c *gin.Context) {
+	c.JSON(200, gin.H{
+		"success": true,
+		"data":    model.GetEnabledModels(),
+	})
+}
+
 func RetrieveModel(c *gin.Context) {
 	modelId := c.Param("model")
 	if aiModel, ok := openAIModelsMap[modelId]; ok {

controller/pricing.go

@@ -2,9 +2,9 @@ package controller
 
 import (
 	"github.com/gin-gonic/gin"
-	"one-api/common"
 	"one-api/model"
 	"one-api/setting"
+	"one-api/setting/operation_setting"
 )
 
 func GetPricing(c *gin.Context) {
@@ -17,7 +17,7 @@ func GetPricing(c *gin.Context) {
 	}
 	var group string
 	if exists {
-		user, err := model.GetUserById(userId.(int), false)
+		user, err := model.GetUserCache(userId.(int))
 		if err == nil {
 			group = user.Group
 		}
@@ -40,7 +40,7 @@ func GetPricing(c *gin.Context) {
 }
 
 func ResetModelRatio(c *gin.Context) {
-	defaultStr := common.DefaultModelRatio2JSONString()
+	defaultStr := operation_setting.DefaultModelRatio2JSONString()
 	err := model.UpdateOption("ModelRatio", defaultStr)
 	if err != nil {
 		c.JSON(200, gin.H{
@@ -49,7 +49,7 @@ func ResetModelRatio(c *gin.Context) {
 		})
 		return
 	}
-	err = common.UpdateModelRatioByJSONString(defaultStr)
+	err = operation_setting.UpdateModelRatioByJSONString(defaultStr)
 	if err != nil {
 		c.JSON(200, gin.H{
 			"success": false,

controller/relay.go

@@ -16,6 +16,7 @@ import (
 	"one-api/relay"
 	"one-api/relay/constant"
 	relayconstant "one-api/relay/constant"
+	"one-api/relay/helper"
 	"one-api/service"
 	"strings"
 )
@@ -24,7 +25,7 @@ func relayHandler(c *gin.Context, relayMode int) *dto.OpenAIErrorWithStatusCode
 	var err *dto.OpenAIErrorWithStatusCode
 	switch relayMode {
 	case relayconstant.RelayModeImagesGenerations:
-		err = relay.ImageHelper(c, relayMode)
+		err = relay.ImageHelper(c)
 	case relayconstant.RelayModeAudioSpeech:
 		fallthrough
 	case relayconstant.RelayModeAudioTranslation:
@@ -33,15 +34,8 @@ func relayHandler(c *gin.Context, relayMode int) *dto.OpenAIErrorWithStatusCode
 		err = relay.AudioHelper(c)
 	case relayconstant.RelayModeRerank:
 		err = relay.RerankHelper(c, relayMode)
-	default:
-		err = relay.TextHelper(c)
-	}
-	return err
-}
-
-func wsHandler(c *gin.Context, ws *websocket.Conn, relayMode int) *dto.OpenAIErrorWithStatusCode {
-	var err *dto.OpenAIErrorWithStatusCode
-	switch relayMode {
+	case relayconstant.RelayModeEmbeddings:
+		err = relay.EmbeddingHelper(c)
 	default:
 		err = relay.TextHelper(c)
 	}
@@ -83,6 +77,7 @@ func Relay(c *gin.Context) {
 
 	if openaiErr != nil {
 		if openaiErr.StatusCode == http.StatusTooManyRequests {
+			common.LogError(c, fmt.Sprintf("origin 429 error: %s", openaiErr.Error.Message))
 			openaiErr.Error.Message = "当前分组上游负载已饱和，请稍后再试"
 		}
 		openaiErr.Error.Message = common.MessageWithRequestId(openaiErr.Error.Message, requestId)
@@ -107,7 +102,7 @@ func WssRelay(c *gin.Context) {
 
 	if err != nil {
 		openaiErr := service.OpenAIErrorWrapper(err, "get_channel_failed", http.StatusInternalServerError)
-		service.WssError(c, ws, openaiErr.Error)
+		helper.WssError(c, ws, openaiErr.Error)
 		return
 	}
 
@@ -149,7 +144,7 @@ func WssRelay(c *gin.Context) {
 			openaiErr.Error.Message = "当前分组上游负载已饱和，请稍后再试"
 		}
 		openaiErr.Error.Message = common.MessageWithRequestId(openaiErr.Error.Message, requestId)
-		service.WssError(c, ws, openaiErr.Error)
+		helper.WssError(c, ws, openaiErr.Error)
 	}
 }
 

controller/task.go

@@ -159,7 +159,7 @@ func updateSunoTaskAll(ctx context.Context, channelId int, taskIds []string, tas
 			} else {
 				quota := task.Quota
 				if quota != 0 {
-					err = model.IncreaseUserQuota(task.UserId, quota)
+					err = model.IncreaseUserQuota(task.UserId, quota, false)
 					if err != nil {
 						common.LogError(ctx, "fail to increase user quota: "+err.Error())
 					}

controller/topup.go

@@ -2,9 +2,6 @@ package controller
 
 import (
 	"fmt"
-	"github.com/Calcium-Ion/go-epay/epay"
-	"github.com/gin-gonic/gin"
-	"github.com/samber/lo"
 	"log"
 	"net/url"
 	"one-api/common"
@@ -14,16 +11,21 @@ import (
 	"strconv"
 	"sync"
 	"time"
+
+	"github.com/Calcium-Ion/go-epay/epay"
+	"github.com/gin-gonic/gin"
+	"github.com/samber/lo"
+	"github.com/shopspring/decimal"
 )
 
 type EpayRequest struct {
-	Amount        int    `json:"amount"`
+	Amount        int64  `json:"amount"`
 	PaymentMethod string `json:"payment_method"`
 	TopUpCode     string `json:"top_up_code"`
 }
 
 type AmountRequest struct {
-	Amount    int    `json:"amount"`
+	Amount    int64  `json:"amount"`
 	TopUpCode string `json:"top_up_code"`
 }
 
@@ -41,25 +43,35 @@ func GetEpayClient() *epay.Client {
 	return withUrl
 }
 
-func getPayMoney(amount float64, group string) float64 {
+func getPayMoney(amount int64, group string) float64 {
+	dAmount := decimal.NewFromInt(amount)
+
 	if !common.DisplayInCurrencyEnabled {
-		amount = amount / common.QuotaPerUnit
+		dQuotaPerUnit := decimal.NewFromFloat(common.QuotaPerUnit)
+		dAmount = dAmount.Div(dQuotaPerUnit)
 	}
-	// 别问为什么用float64，问就是这么点钱没必要
+
 	topupGroupRatio := common.GetTopupGroupRatio(group)
 	if topupGroupRatio == 0 {
 		topupGroupRatio = 1
 	}
-	payMoney := amount * setting.Price * topupGroupRatio
-	return payMoney
+
+	dTopupGroupRatio := decimal.NewFromFloat(topupGroupRatio)
+	dPrice := decimal.NewFromFloat(setting.Price)
+
+	payMoney := dAmount.Mul(dPrice).Mul(dTopupGroupRatio)
+
+	return payMoney.InexactFloat64()
 }
 
-func getMinTopup() int {
+func getMinTopup() int64 {
 	minTopup := setting.MinTopUp
 	if !common.DisplayInCurrencyEnabled {
-		minTopup = minTopup * int(common.QuotaPerUnit)
+		dMinTopup := decimal.NewFromInt(int64(minTopup))
+		dQuotaPerUnit := decimal.NewFromFloat(common.QuotaPerUnit)
+		minTopup = int(dMinTopup.Mul(dQuotaPerUnit).IntPart())
 	}
-	return minTopup
+	return int64(minTopup)
 }
 
 func RequestEpay(c *gin.Context) {
@@ -80,7 +92,7 @@ func RequestEpay(c *gin.Context) {
 		c.JSON(200, gin.H{"message": "error", "data": "获取用户分组失败"})
 		return
 	}
-	payMoney := getPayMoney(float64(req.Amount), group)
+	payMoney := getPayMoney(req.Amount, group)
 	if payMoney < 0.01 {
 		c.JSON(200, gin.H{"message": "error", "data": "充值金额过低"})
 		return
@@ -118,7 +130,9 @@ func RequestEpay(c *gin.Context) {
 	}
 	amount := req.Amount
 	if !common.DisplayInCurrencyEnabled {
-		amount = amount / int(common.QuotaPerUnit)
+		dAmount := decimal.NewFromInt(int64(amount))
+		dQuotaPerUnit := decimal.NewFromFloat(common.QuotaPerUnit)
+		amount = dAmount.Div(dQuotaPerUnit).IntPart()
 	}
 	topUp := &model.TopUp{
 		UserId:     id,
@@ -210,13 +224,16 @@ func EpayNotify(c *gin.Context) {
 			}
 			//user, _ := model.GetUserById(topUp.UserId, false)
 			//user.Quota += topUp.Amount * 500000
-			err = model.IncreaseUserQuota(topUp.UserId, topUp.Amount*int(common.QuotaPerUnit))
+			dAmount := decimal.NewFromInt(int64(topUp.Amount))
+			dQuotaPerUnit := decimal.NewFromFloat(common.QuotaPerUnit)
+			quotaToAdd := int(dAmount.Mul(dQuotaPerUnit).IntPart())
+			err = model.IncreaseUserQuota(topUp.UserId, quotaToAdd, true)
 			if err != nil {
 				log.Printf("易支付回调更新用户失败: %v", topUp)
 				return
 			}
 			log.Printf("易支付回调更新用户成功 %v", topUp)
-			model.RecordLog(topUp.UserId, model.LogTypeTopup, fmt.Sprintf("使用在线充值成功，充值金额: %v，支付金额：%f", common.LogQuota(topUp.Amount*int(common.QuotaPerUnit)), topUp.Money))
+			model.RecordLog(topUp.UserId, model.LogTypeTopup, fmt.Sprintf("使用在线充值成功，充值金额: %v，支付金额：%f", common.LogQuota(quotaToAdd), topUp.Money))
 		}
 	} else {
 		log.Printf("易支付异常回调: %v", verifyInfo)
@@ -241,7 +258,7 @@ func RequestAmount(c *gin.Context) {
 		c.JSON(200, gin.H{"message": "error", "data": "获取用户分组失败"})
 		return
 	}
-	payMoney := getPayMoney(float64(req.Amount), group)
+	payMoney := getPayMoney(req.Amount, group)
 	if payMoney <= 0.01 {
 		c.JSON(200, gin.H{"message": "error", "data": "充值金额过低"})
 		return

controller/user.go

@@ -4,6 +4,7 @@ import (
 	"encoding/json"
 	"fmt"
 	"net/http"
+	"net/url"
 	"one-api/common"
 	"one-api/model"
 	"one-api/setting"
@@ -471,7 +472,7 @@ func GetUserModels(c *gin.Context) {
 	if err != nil {
 		id = c.GetInt("id")
 	}
-	user, err := model.GetUserById(id, true)
+	user, err := model.GetUserCache(id)
 	if err != nil {
 		c.JSON(http.StatusOK, gin.H{
 			"success": false,
@@ -846,9 +847,10 @@ func EmailBind(c *gin.Context) {
 		})
 		return
 	}
-	id := c.GetInt("id")
+	session := sessions.Default(c)
+	id := session.Get("id")
 	user := model.User{
-		Id: id,
+		Id: id.(int),
 	}
 	err := user.FillUserById()
 	if err != nil {
@@ -868,9 +870,6 @@ func EmailBind(c *gin.Context) {
 		})
 		return
 	}
-	if user.Role == common.RoleRootUser {
-		common.RootUserEmail = email
-	}
 	c.JSON(http.StatusOK, gin.H{
 		"success": true,
 		"message": "",
@@ -912,3 +911,115 @@ func TopUp(c *gin.Context) {
 	})
 	return
 }
+
+type UpdateUserSettingRequest struct {
+	QuotaWarningType      string  `json:"notify_type"`
+	QuotaWarningThreshold float64 `json:"quota_warning_threshold"`
+	WebhookUrl            string  `json:"webhook_url,omitempty"`
+	WebhookSecret         string  `json:"webhook_secret,omitempty"`
+	NotificationEmail     string  `json:"notification_email,omitempty"`
+}
+
+func UpdateUserSetting(c *gin.Context) {
+	var req UpdateUserSettingRequest
+	if err := c.ShouldBindJSON(&req); err != nil {
+		c.JSON(http.StatusOK, gin.H{
+			"success": false,
+			"message": "无效的参数",
+		})
+		return
+	}
+
+	// 验证预警类型
+	if req.QuotaWarningType != constant.NotifyTypeEmail && req.QuotaWarningType != constant.NotifyTypeWebhook {
+		c.JSON(http.StatusOK, gin.H{
+			"success": false,
+			"message": "无效的预警类型",
+		})
+		return
+	}
+
+	// 验证预警阈值
+	if req.QuotaWarningThreshold <= 0 {
+		c.JSON(http.StatusOK, gin.H{
+			"success": false,
+			"message": "预警阈值必须大于0",
+		})
+		return
+	}
+
+	// 如果是webhook类型,验证webhook地址
+	if req.QuotaWarningType == constant.NotifyTypeWebhook {
+		if req.WebhookUrl == "" {
+			c.JSON(http.StatusOK, gin.H{
+				"success": false,
+				"message": "Webhook地址不能为空",
+			})
+			return
+		}
+		// 验证URL格式
+		if _, err := url.ParseRequestURI(req.WebhookUrl); err != nil {
+			c.JSON(http.StatusOK, gin.H{
+				"success": false,
+				"message": "无效的Webhook地址",
+			})
+			return
+		}
+	}
+
+	// 如果是邮件类型，验证邮箱地址
+	if req.QuotaWarningType == constant.NotifyTypeEmail && req.NotificationEmail != "" {
+		// 验证邮箱格式
+		if !strings.Contains(req.NotificationEmail, "@") {
+			c.JSON(http.StatusOK, gin.H{
+				"success": false,
+				"message": "无效的邮箱地址",
+			})
+			return
+		}
+	}
+
+	userId := c.GetInt("id")
+	user, err := model.GetUserById(userId, true)
+	if err != nil {
+		c.JSON(http.StatusOK, gin.H{
+			"success": false,
+			"message": err.Error(),
+		})
+		return
+	}
+
+	// 构建设置
+	settings := map[string]interface{}{
+		constant.UserSettingNotifyType:            req.QuotaWarningType,
+		constant.UserSettingQuotaWarningThreshold: req.QuotaWarningThreshold,
+	}
+
+	// 如果是webhook类型,添加webhook相关设置
+	if req.QuotaWarningType == constant.NotifyTypeWebhook {
+		settings[constant.UserSettingWebhookUrl] = req.WebhookUrl
+		if req.WebhookSecret != "" {
+			settings[constant.UserSettingWebhookSecret] = req.WebhookSecret
+		}
+	}
+
+	// 如果提供了通知邮箱，添加到设置中
+	if req.QuotaWarningType == constant.NotifyTypeEmail && req.NotificationEmail != "" {
+		settings[constant.UserSettingNotificationEmail] = req.NotificationEmail
+	}
+
+	// 更新用户设置
+	user.SetSetting(settings)
+	if err := user.Update(false); err != nil {
+		c.JSON(http.StatusOK, gin.H{
+			"success": false,
+			"message": "更新设置失败: " + err.Error(),
+		})
+		return
+	}
+
+	c.JSON(http.StatusOK, gin.H{
+		"success": true,
+		"message": "设置已更新",
+	})
+}

controller/wechat.go

@@ -4,6 +4,7 @@ import (
 	"encoding/json"
 	"errors"
 	"fmt"
+	"github.com/gin-contrib/sessions"
 	"github.com/gin-gonic/gin"
 	"net/http"
 	"one-api/common"
@@ -142,9 +143,10 @@ func WeChatBind(c *gin.Context) {
 		})
 		return
 	}
-	id := c.GetInt("id")
+	session := sessions.Default(c)
+	id := session.Get("id")
 	user := model.User{
-		Id: id,
+		Id: id.(int),
 	}
 	err = user.FillUserById()
 	if err != nil {

docker-compose.yml

@@ -24,7 +24,7 @@ services:
       - redis
       - mysql
     healthcheck:
-      test: [ "CMD-SHELL", "wget -q -O - http://localhost:3000/api/status | grep -o '\"success\":\\s*true' | awk -F: '{print $2}'" ]
+      test: ["CMD-SHELL", "wget -q -O - http://localhost:3000/api/status | grep -o '\"success\":\\s*true' | awk -F: '{print $$2}'"]
       interval: 30s
       timeout: 10s
       retries: 3

docs/channel/other_setting.md

@@ -0,0 +1,33 @@
+# 渠道而外设置说明
+
+该配置用于设置一些额外的渠道参数，可以通过 JSON 对象进行配置。主要包含以下两个设置项：
+
+1. force_format
+    - 用于标识是否对数据进行强制格式化为 OpenAI 格式
+    - 类型为布尔值，设置为 true 时启用强制格式化
+
+2. proxy
+    - 用于配置网络代理
+    - 类型为字符串，填写代理地址（例如 socks5 协议的代理地址）
+
+3. thinking_to_content
+   - 用于标识是否将思考内容`reasoning_conetnt`转换为`<think>`标签拼接到内容中返回
+   - 类型为布尔值，设置为 true 时启用思考内容转换
+
+--------------------------------------------------------------
+
+## JSON 格式示例
+
+以下是一个示例配置，启用强制格式化并设置了代理地址：
+
+```json
+{
+    "force_format": true,
+   "thinking_to_content": true,
+    "proxy": "socks5://xxxxxxx"
+}
+```
+
+--------------------------------------------------------------
+
+通过调整上述 JSON 配置中的值，可以灵活控制渠道的额外行为，比如是否进行格式化以及使用特定的网络代理。

dto/embedding.go

@@ -0,0 +1,57 @@
+package dto
+
+type EmbeddingOptions struct {
+	Seed             int      `json:"seed,omitempty"`
+	Temperature      *float64 `json:"temperature,omitempty"`
+	TopK             int      `json:"top_k,omitempty"`
+	TopP             *float64 `json:"top_p,omitempty"`
+	FrequencyPenalty *float64 `json:"frequency_penalty,omitempty"`
+	PresencePenalty  *float64 `json:"presence_penalty,omitempty"`
+	NumPredict       int      `json:"num_predict,omitempty"`
+	NumCtx           int      `json:"num_ctx,omitempty"`
+}
+
+type EmbeddingRequest struct {
+	Model            string   `json:"model"`
+	Input            any      `json:"input"`
+	EncodingFormat   string   `json:"encoding_format,omitempty"`
+	Dimensions       int      `json:"dimensions,omitempty"`
+	User             string   `json:"user,omitempty"`
+	Seed             float64  `json:"seed,omitempty"`
+	Temperature      *float64 `json:"temperature,omitempty"`
+	TopP             float64  `json:"top_p,omitempty"`
+	FrequencyPenalty float64  `json:"frequency_penalty,omitempty"`
+	PresencePenalty  float64  `json:"presence_penalty,omitempty"`
+}
+
+func (r EmbeddingRequest) ParseInput() []string {
+	if r.Input == nil {
+		return nil
+	}
+	var input []string
+	switch r.Input.(type) {
+	case string:
+		input = []string{r.Input.(string)}
+	case []any:
+		input = make([]string, 0, len(r.Input.([]any)))
+		for _, item := range r.Input.([]any) {
+			if str, ok := item.(string); ok {
+				input = append(input, str)
+			}
+		}
+	}
+	return input
+}
+
+type EmbeddingResponseItem struct {
+	Object    string    `json:"object"`
+	Index     int       `json:"index"`
+	Embedding []float64 `json:"embedding"`
+}
+
+type EmbeddingResponse struct {
+	Object string                  `json:"object"`
+	Data   []EmbeddingResponseItem `json:"data"`
+	Model  string                  `json:"model"`
+	Usage  `json:"usage"`
+}

dto/notify.go

@@ -0,0 +1,25 @@
+package dto
+
+type Notify struct {
+	Type    string        `json:"type"`
+	Title   string        `json:"title"`
+	Content string        `json:"content"`
+	Values  []interface{} `json:"values"`
+}
+
+const ContentValueParam = "{{value}}"
+
+const (
+	NotifyTypeQuotaExceed   = "quota_exceed"
+	NotifyTypeChannelUpdate = "channel_update"
+	NotifyTypeChannelTest   = "channel_test"
+)
+
+func NewNotify(t string, title string, content string, values []interface{}) Notify {
+	return Notify{
+		Type:    t,
+		Title:   title,
+		Content: content,
+		Values:  values,
+	}
+}

dto/openai_request.go

@@ -1,6 +1,9 @@
 package dto
 
-import "encoding/json"
+import (
+	"encoding/json"
+	"strings"
+)
 
 type ResponseFormat struct {
 	Type       string            `json:"type,omitempty"`
@@ -15,47 +18,52 @@ type FormatJsonSchema struct {
 }
 
 type GeneralOpenAIRequest struct {
-	Model               string          `json:"model,omitempty"`
-	Messages            []Message       `json:"messages,omitempty"`
-	Prompt              any             `json:"prompt,omitempty"`
-	Stream              bool            `json:"stream,omitempty"`
-	StreamOptions       *StreamOptions  `json:"stream_options,omitempty"`
-	MaxTokens           uint            `json:"max_tokens,omitempty"`
-	MaxCompletionTokens uint            `json:"max_completion_tokens,omitempty"`
-	ReasoningEffort     string          `json:"reasoning_effort,omitempty"`
-	Temperature         *float64        `json:"temperature,omitempty"`
-	TopP                float64         `json:"top_p,omitempty"`
-	TopK                int             `json:"top_k,omitempty"`
-	Stop                any             `json:"stop,omitempty"`
-	N                   int             `json:"n,omitempty"`
-	Input               any             `json:"input,omitempty"`
-	Instruction         string          `json:"instruction,omitempty"`
-	Size                string          `json:"size,omitempty"`
-	Functions           any             `json:"functions,omitempty"`
-	FrequencyPenalty    float64         `json:"frequency_penalty,omitempty"`
-	PresencePenalty     float64         `json:"presence_penalty,omitempty"`
-	ResponseFormat      *ResponseFormat `json:"response_format,omitempty"`
-	EncodingFormat      any             `json:"encoding_format,omitempty"`
-	Seed                float64         `json:"seed,omitempty"`
-	Tools               []ToolCall      `json:"tools,omitempty"`
-	ToolChoice          any             `json:"tool_choice,omitempty"`
-	User                string          `json:"user,omitempty"`
-	LogProbs            bool            `json:"logprobs,omitempty"`
-	TopLogProbs         int             `json:"top_logprobs,omitempty"`
-	Dimensions          int             `json:"dimensions,omitempty"`
-	Modalities          any             `json:"modalities,omitempty"`
-	Audio               any             `json:"audio,omitempty"`
+	Model               string            `json:"model,omitempty"`
+	Messages            []Message         `json:"messages,omitempty"`
+	Prompt              any               `json:"prompt,omitempty"`
+	Prefix              any               `json:"prefix,omitempty"`
+	Suffix              any               `json:"suffix,omitempty"`
+	Stream              bool              `json:"stream,omitempty"`
+	StreamOptions       *StreamOptions    `json:"stream_options,omitempty"`
+	MaxTokens           uint              `json:"max_tokens,omitempty"`
+	MaxCompletionTokens uint              `json:"max_completion_tokens,omitempty"`
+	ReasoningEffort     string            `json:"reasoning_effort,omitempty"`
+	Temperature         *float64          `json:"temperature,omitempty"`
+	TopP                float64           `json:"top_p,omitempty"`
+	TopK                int               `json:"top_k,omitempty"`
+	Stop                any               `json:"stop,omitempty"`
+	N                   int               `json:"n,omitempty"`
+	Input               any               `json:"input,omitempty"`
+	Instruction         string            `json:"instruction,omitempty"`
+	Size                string            `json:"size,omitempty"`
+	Functions           any               `json:"functions,omitempty"`
+	FrequencyPenalty    float64           `json:"frequency_penalty,omitempty"`
+	PresencePenalty     float64           `json:"presence_penalty,omitempty"`
+	ResponseFormat      *ResponseFormat   `json:"response_format,omitempty"`
+	EncodingFormat      any               `json:"encoding_format,omitempty"`
+	Seed                float64           `json:"seed,omitempty"`
+	Tools               []ToolCallRequest `json:"tools,omitempty"`
+	ToolChoice          any               `json:"tool_choice,omitempty"`
+	User                string            `json:"user,omitempty"`
+	LogProbs            bool              `json:"logprobs,omitempty"`
+	TopLogProbs         int               `json:"top_logprobs,omitempty"`
+	Dimensions          int               `json:"dimensions,omitempty"`
+	Modalities          any               `json:"modalities,omitempty"`
+	Audio               any               `json:"audio,omitempty"`
+	ExtraBody           any               `json:"extra_body,omitempty"`
 }
 
-type OpenAITools struct {
-	Type     string         `json:"type"`
-	Function OpenAIFunction `json:"function"`
+type ToolCallRequest struct {
+	ID       string          `json:"id,omitempty"`
+	Type     string          `json:"type"`
+	Function FunctionRequest `json:"function"`
 }
 
-type OpenAIFunction struct {
+type FunctionRequest struct {
 	Description string `json:"description,omitempty"`
 	Name        string `json:"name"`
 	Parameters  any    `json:"parameters,omitempty"`
+	Arguments   string `json:"arguments,omitempty"`
 }
 
 type StreamOptions struct {
@@ -86,16 +94,21 @@ func (r GeneralOpenAIRequest) ParseInput() []string {
 }
 
 type Message struct {
-	Role       string          `json:"role"`
-	Content    json.RawMessage `json:"content"`
-	Name       *string         `json:"name,omitempty"`
-	ToolCalls  json.RawMessage `json:"tool_calls,omitempty"`
-	ToolCallId string          `json:"tool_call_id,omitempty"`
+	Role                string          `json:"role"`
+	Content             json.RawMessage `json:"content"`
+	Name                *string         `json:"name,omitempty"`
+	Prefix              *bool           `json:"prefix,omitempty"`
+	ReasoningContent    string          `json:"reasoning_content,omitempty"`
+	Reasoning           string          `json:"reasoning,omitempty"`
+	ToolCalls           json.RawMessage `json:"tool_calls,omitempty"`
+	ToolCallId          string          `json:"tool_call_id,omitempty"`
+	parsedContent       []MediaContent
+	parsedStringContent *string
 }
 
 type MediaContent struct {
 	Type       string `json:"type"`
-	Text       string `json:"text"`
+	Text       string `json:"text,omitempty"`
 	ImageUrl   any    `json:"image_url,omitempty"`
 	InputAudio any    `json:"input_audio,omitempty"`
 }
@@ -116,11 +129,22 @@ const (
 	ContentTypeInputAudio = "input_audio"
 )
 
-func (m *Message) ParseToolCalls() []ToolCall {
+func (m *Message) GetPrefix() bool {
+	if m.Prefix == nil {
+		return false
+	}
+	return *m.Prefix
+}
+
+func (m *Message) SetPrefix(prefix bool) {
+	m.Prefix = &prefix
+}
+
+func (m *Message) ParseToolCalls() []ToolCallRequest {
 	if m.ToolCalls == nil {
 		return nil
 	}
-	var toolCalls []ToolCall
+	var toolCalls []ToolCallRequest
 	if err := json.Unmarshal(m.ToolCalls, &toolCalls); err == nil {
 		return toolCalls
 	}
@@ -133,88 +157,139 @@ func (m *Message) SetToolCalls(toolCalls any) {
 }
 
 func (m *Message) StringContent() string {
+	if m.parsedStringContent != nil {
+		return *m.parsedStringContent
+	}
+
 	var stringContent string
 	if err := json.Unmarshal(m.Content, &stringContent); err == nil {
+		m.parsedStringContent = &stringContent
 		return stringContent
 	}
-	return string(m.Content)
+
+	contentStr := new(strings.Builder)
+	arrayContent := m.ParseContent()
+	for _, content := range arrayContent {
+		if content.Type == ContentTypeText {
+			contentStr.WriteString(content.Text)
+		}
+	}
+	stringContent = contentStr.String()
+	m.parsedStringContent = &stringContent
+
+	return stringContent
 }
 
 func (m *Message) SetStringContent(content string) {
 	jsonContent, _ := json.Marshal(content)
 	m.Content = jsonContent
+	m.parsedStringContent = &content
+	m.parsedContent = nil
+}
+
+func (m *Message) SetMediaContent(content []MediaContent) {
+	jsonContent, _ := json.Marshal(content)
+	m.Content = jsonContent
+	m.parsedContent = nil
+	m.parsedStringContent = nil
 }
 
 func (m *Message) IsStringContent() bool {
+	if m.parsedStringContent != nil {
+		return true
+	}
 	var stringContent string
 	if err := json.Unmarshal(m.Content, &stringContent); err == nil {
+		m.parsedStringContent = &stringContent
 		return true
 	}
 	return false
 }
 
 func (m *Message) ParseContent() []MediaContent {
+	if m.parsedContent != nil {
+		return m.parsedContent
+	}
+
 	var contentList []MediaContent
+
+	// 先尝试解析为字符串
 	var stringContent string
 	if err := json.Unmarshal(m.Content, &stringContent); err == nil {
-		contentList = append(contentList, MediaContent{
+		contentList = []MediaContent{{
 			Type: ContentTypeText,
 			Text: stringContent,
-		})
+		}}
+		m.parsedContent = contentList
 		return contentList
 	}
-	var arrayContent []json.RawMessage
+
+	// 尝试解析为数组
+	var arrayContent []map[string]interface{}
 	if err := json.Unmarshal(m.Content, &arrayContent); err == nil {
 		for _, contentItem := range arrayContent {
-			var contentMap map[string]any
-			if err := json.Unmarshal(contentItem, &contentMap); err != nil {
+			contentType, ok := contentItem["type"].(string)
+			if !ok {
 				continue
 			}
-			switch contentMap["type"] {
+
+			switch contentType {
 			case ContentTypeText:
-				if subStr, ok := contentMap["text"].(string); ok {
+				if text, ok := contentItem["text"].(string); ok {
 					contentList = append(contentList, MediaContent{
 						Type: ContentTypeText,
-						Text: subStr,
+						Text: text,
 					})
 				}
+
 			case ContentTypeImageURL:
-				if subObj, ok := contentMap["image_url"].(map[string]any); ok {
-					detail, ok := subObj["detail"]
-					if ok {
-						subObj["detail"] = detail.(string)
-					} else {
-						subObj["detail"] = "high"
-					}
+				imageUrl := contentItem["image_url"]
+				switch v := imageUrl.(type) {
+				case string:
 					contentList = append(contentList, MediaContent{
 						Type: ContentTypeImageURL,
 						ImageUrl: MessageImageUrl{
-							Url:    subObj["url"].(string),
-							Detail: subObj["detail"].(string),
-						},
-					})
-				} else if url, ok := contentMap["image_url"].(string); ok {
-					contentList = append(contentList, MediaContent{
-						Type: ContentTypeImageURL,
-						ImageUrl: MessageImageUrl{
-							Url:    url,
+							Url:    v,
 							Detail: "high",
 						},
 					})
+				case map[string]interface{}:
+					url, ok1 := v["url"].(string)
+					detail, ok2 := v["detail"].(string)
+					if !ok2 {
+						detail = "high"
+					}
+					if ok1 {
+						contentList = append(contentList, MediaContent{
+							Type: ContentTypeImageURL,
+							ImageUrl: MessageImageUrl{
+								Url:    url,
+								Detail: detail,
+							},
+						})
+					}
 				}
+
 			case ContentTypeInputAudio:
-				if subObj, ok := contentMap["input_audio"].(map[string]any); ok {
-					contentList = append(contentList, MediaContent{
-						Type: ContentTypeInputAudio,
-						InputAudio: MessageInputAudio{
-							Data:   subObj["data"].(string),
-							Format: subObj["format"].(string),
-						},
-					})
+				if audioData, ok := contentItem["input_audio"].(map[string]interface{}); ok {
+					data, ok1 := audioData["data"].(string)
+					format, ok2 := audioData["format"].(string)
+					if ok1 && ok2 {
+						contentList = append(contentList, MediaContent{
+							Type: ContentTypeInputAudio,
+							InputAudio: MessageInputAudio{
+								Data:   data,
+								Format: format,
+							},
+						})
+					}
 				}
 			}
 		}
-		return contentList
 	}
-	return nil
+
+	if len(contentList) > 0 {
+		m.parsedContent = contentList
+	}
+	return contentList
 }

dto/openai_response.go

@@ -62,9 +62,11 @@ type ChatCompletionsStreamResponseChoice struct {
 }
 
 type ChatCompletionsStreamResponseChoiceDelta struct {
-	Content   *string    `json:"content,omitempty"`
-	Role      string     `json:"role,omitempty"`
-	ToolCalls []ToolCall `json:"tool_calls,omitempty"`
+	Content          *string            `json:"content,omitempty"`
+	ReasoningContent *string            `json:"reasoning_content,omitempty"`
+	Reasoning        *string            `json:"reasoning,omitempty"`
+	Role             string             `json:"role,omitempty"`
+	ToolCalls        []ToolCallResponse `json:"tool_calls,omitempty"`
 }
 
 func (c *ChatCompletionsStreamResponseChoiceDelta) SetContentString(s string) {
@@ -78,24 +80,39 @@ func (c *ChatCompletionsStreamResponseChoiceDelta) GetContentString() string {
 	return *c.Content
 }
 
-type ToolCall struct {
-	// Index is not nil only in chat completion chunk object
-	Index    *int         `json:"index,omitempty"`
-	ID       string       `json:"id"`
-	Type     any          `json:"type"`
-	Function FunctionCall `json:"function"`
+func (c *ChatCompletionsStreamResponseChoiceDelta) GetReasoningContent() string {
+	if c.ReasoningContent == nil && c.Reasoning == nil {
+		return ""
+	}
+	if c.ReasoningContent != nil {
+		return *c.ReasoningContent
+	}
+	return *c.Reasoning
 }
 
-func (c *ToolCall) SetIndex(i int) {
+func (c *ChatCompletionsStreamResponseChoiceDelta) SetReasoningContent(s string) {
+	c.ReasoningContent = &s
+	c.Reasoning = &s
+}
+
+type ToolCallResponse struct {
+	// Index is not nil only in chat completion chunk object
+	Index    *int             `json:"index,omitempty"`
+	ID       string           `json:"id,omitempty"`
+	Type     any              `json:"type"`
+	Function FunctionResponse `json:"function"`
+}
+
+func (c *ToolCallResponse) SetIndex(i int) {
 	c.Index = &i
 }
 
-type FunctionCall struct {
+type FunctionResponse struct {
 	Description string `json:"description,omitempty"`
 	Name        string `json:"name,omitempty"`
 	// call function with arguments in JSON format
 	Parameters any    `json:"parameters,omitempty"` // request
-	Arguments  string `json:"arguments,omitempty"`
+	Arguments  string `json:"arguments"`            // response
 }
 
 type ChatCompletionsStreamResponse struct {
@@ -108,6 +125,20 @@ type ChatCompletionsStreamResponse struct {
 	Usage             *Usage                                `json:"usage"`
 }
 
+func (c *ChatCompletionsStreamResponse) Copy() *ChatCompletionsStreamResponse {
+	choices := make([]ChatCompletionsStreamResponseChoice, len(c.Choices))
+	copy(choices, c.Choices)
+	return &ChatCompletionsStreamResponse{
+		Id:                c.Id,
+		Object:            c.Object,
+		Created:           c.Created,
+		Model:             c.Model,
+		SystemFingerprint: c.SystemFingerprint,
+		Choices:           choices,
+		Usage:             c.Usage,
+	}
+}
+
 func (c *ChatCompletionsStreamResponse) GetSystemFingerprint() string {
 	if c.SystemFingerprint == nil {
 		return ""
@@ -135,6 +166,7 @@ type Usage struct {
 	PromptTokens           int                `json:"prompt_tokens"`
 	CompletionTokens       int                `json:"completion_tokens"`
 	TotalTokens            int                `json:"total_tokens"`
+	PromptCacheHitTokens   int                `json:"prompt_cache_hit_tokens,omitempty"`
 	PromptTokensDetails    InputTokenDetails  `json:"prompt_tokens_details"`
 	CompletionTokenDetails OutputTokenDetails `json:"completion_tokens_details"`
 }

go.mod

@@ -16,23 +16,24 @@ require (
 	github.com/gin-contrib/sessions v0.0.5
 	github.com/gin-contrib/static v0.0.1
 	github.com/gin-gonic/gin v1.9.1
+	github.com/glebarez/sqlite v1.9.0
 	github.com/go-playground/validator/v10 v10.20.0
 	github.com/go-redis/redis/v8 v8.11.5
 	github.com/golang-jwt/jwt v3.2.2+incompatible
 	github.com/google/uuid v1.6.0
 	github.com/gorilla/websocket v1.5.0
-	github.com/jinzhu/copier v0.4.0
 	github.com/joho/godotenv v1.5.1
 	github.com/pkg/errors v0.9.1
 	github.com/pkoukk/tiktoken-go v0.1.7
 	github.com/samber/lo v1.39.0
 	github.com/shirou/gopsutil v3.21.11+incompatible
+	github.com/shopspring/decimal v1.4.0
 	golang.org/x/crypto v0.27.0
 	golang.org/x/image v0.23.0
+	golang.org/x/net v0.28.0
 	gorm.io/driver/mysql v1.4.3
 	gorm.io/driver/postgres v1.5.2
-	gorm.io/driver/sqlite v1.4.3
-	gorm.io/gorm v1.25.0
+	gorm.io/gorm v1.25.2
 )
 
 require (
@@ -48,12 +49,14 @@ require (
 	github.com/cloudwego/iasm v0.2.0 // indirect
 	github.com/dgryski/go-rendezvous v0.0.0-20200823014737-9f7001d12a5f // indirect
 	github.com/dlclark/regexp2 v1.11.0 // indirect
+	github.com/dustin/go-humanize v1.0.1 // indirect
 	github.com/gabriel-vasile/mimetype v1.4.3 // indirect
 	github.com/gin-contrib/sse v0.1.0 // indirect
+	github.com/glebarez/go-sqlite v1.21.2 // indirect
 	github.com/go-ole/go-ole v1.2.6 // indirect
 	github.com/go-playground/locales v0.14.1 // indirect
 	github.com/go-playground/universal-translator v0.18.1 // indirect
-	github.com/go-sql-driver/mysql v1.6.0 // indirect
+	github.com/go-sql-driver/mysql v1.7.0 // indirect
 	github.com/goccy/go-json v0.10.2 // indirect
 	github.com/google/go-cmp v0.6.0 // indirect
 	github.com/gorilla/context v1.1.1 // indirect
@@ -69,11 +72,11 @@ require (
 	github.com/klauspost/cpuid/v2 v2.2.9 // indirect
 	github.com/leodido/go-urn v1.4.0 // indirect
 	github.com/mattn/go-isatty v0.0.20 // indirect
-	github.com/mattn/go-sqlite3 v2.0.3+incompatible // indirect
 	github.com/mitchellh/mapstructure v1.5.0 // indirect
 	github.com/modern-go/concurrent v0.0.0-20180306012644-bacd9c7ef1dd // indirect
 	github.com/modern-go/reflect2 v1.0.2 // indirect
 	github.com/pelletier/go-toml/v2 v2.2.1 // indirect
+	github.com/remyoudompheng/bigfft v0.0.0-20230129092748-24d4a6f8daec // indirect
 	github.com/tklauser/go-sysconf v0.3.12 // indirect
 	github.com/tklauser/numcpus v0.6.1 // indirect
 	github.com/twitchyliquid64/golang-asm v0.15.1 // indirect
@@ -81,10 +84,13 @@ require (
 	github.com/yusufpapurcu/wmi v1.2.3 // indirect
 	golang.org/x/arch v0.12.0 // indirect
 	golang.org/x/exp v0.0.0-20240404231335-c0f41cb1a7a0 // indirect
-	golang.org/x/net v0.28.0 // indirect
 	golang.org/x/sync v0.10.0 // indirect
 	golang.org/x/sys v0.27.0 // indirect
 	golang.org/x/text v0.21.0 // indirect
 	google.golang.org/protobuf v1.34.2 // indirect
 	gopkg.in/yaml.v3 v3.0.1 // indirect
+	modernc.org/libc v1.22.5 // indirect
+	modernc.org/mathutil v1.5.0 // indirect
+	modernc.org/memory v1.5.0 // indirect
+	modernc.org/sqlite v1.23.1 // indirect
 )

go.sum

@@ -40,6 +40,8 @@ github.com/dgryski/go-rendezvous v0.0.0-20200823014737-9f7001d12a5f h1:lO4WD4F/r
 github.com/dgryski/go-rendezvous v0.0.0-20200823014737-9f7001d12a5f/go.mod h1:cuUVRXasLTGF7a8hSLbxyZXjz+1KgoB3wDUb6vlszIc=
 github.com/dlclark/regexp2 v1.11.0 h1:G/nrcoOa7ZXlpoa/91N3X7mM3r8eIlMBBJZvsz/mxKI=
 github.com/dlclark/regexp2 v1.11.0/go.mod h1:DHkYz0B9wPfa6wondMfaivmHpzrQ3v9q8cnmRbL6yW8=
+github.com/dustin/go-humanize v1.0.1 h1:GzkhY7T5VNhEkwH0PVJgjz+fX1rhBrR7pRT3mDkpeCY=
+github.com/dustin/go-humanize v1.0.1/go.mod h1:Mu1zIs6XwVuF/gI1OepvI0qD18qycQx+mFykh5fBlto=
 github.com/fsnotify/fsnotify v1.4.9 h1:hsms1Qyu0jgnwNXIxa+/V/PDsU6CfLf6CNO8H7IWoS4=
 github.com/fsnotify/fsnotify v1.4.9/go.mod h1:znqG4EE+3YCdAaPaxE2ZRY/06pZUdp0tY4IgpuI1SZQ=
 github.com/gabriel-vasile/mimetype v1.4.3 h1:in2uUcidCuFcDKtdcBxlR0rJ1+fsokWf+uqxgUFjbI0=
@@ -58,6 +60,10 @@ github.com/gin-gonic/gin v1.6.3/go.mod h1:75u5sXoLsGZoRN5Sgbi1eraJ4GU3++wFwWzhwv
 github.com/gin-gonic/gin v1.8.1/go.mod h1:ji8BvRH1azfM+SYow9zQ6SZMvR8qOMZHmsCuWR9tTTk=
 github.com/gin-gonic/gin v1.9.1 h1:4idEAncQnU5cB7BeOkPtxjfCSye0AAm1R0RVIqJ+Jmg=
 github.com/gin-gonic/gin v1.9.1/go.mod h1:hPrL7YrpYKXt5YId3A/Tnip5kqbEAP+KLuI3SUcPTeU=
+github.com/glebarez/go-sqlite v1.21.2 h1:3a6LFC4sKahUunAmynQKLZceZCOzUthkRkEAl9gAXWo=
+github.com/glebarez/go-sqlite v1.21.2/go.mod h1:sfxdZyhQjTM2Wry3gVYWaW072Ri1WMdWJi0k6+3382k=
+github.com/glebarez/sqlite v1.9.0 h1:Aj6bPA12ZEx5GbSF6XADmCkYXlljPNUY+Zf1EQxynXs=
+github.com/glebarez/sqlite v1.9.0/go.mod h1:YBYCoyupOao60lzp1MVBLEjZfgkq0tdB1voAQ09K9zw=
 github.com/go-ole/go-ole v1.2.6 h1:/Fpf6oFPoeFik9ty7siob0G6Ke8QvQEuVcuChpwXzpY=
 github.com/go-ole/go-ole v1.2.6/go.mod h1:pprOEPIfldk/42T2oK7lQ4v4JSDwmV0As9GaiUsvbm0=
 github.com/go-playground/assert/v2 v2.0.1/go.mod h1:VDjEfimB/XKnb+ZQfWdccd7VUvScMdVu0Titje2rxJ4=
@@ -77,8 +83,9 @@ github.com/go-playground/validator/v10 v10.20.0 h1:K9ISHbSaI0lyB2eWMPJo+kOS/FBEx
 github.com/go-playground/validator/v10 v10.20.0/go.mod h1:dbuPbCMFw/DrkbEynArYaCwl3amGuJotoKCe95atGMM=
 github.com/go-redis/redis/v8 v8.11.5 h1:AcZZR7igkdvfVmQTPnu9WE37LRrO/YrBH5zWyjDC0oI=
 github.com/go-redis/redis/v8 v8.11.5/go.mod h1:gREzHqY1hg6oD9ngVRbLStwAWKhA0FEgq8Jd4h5lpwo=
-github.com/go-sql-driver/mysql v1.6.0 h1:BCTh4TKNUYmOmMUcQ3IipzF5prigylS7XXjEkfCHuOE=
 github.com/go-sql-driver/mysql v1.6.0/go.mod h1:DCzpHaOWr8IXmIStZouvnhqoel9Qv2LBy8hT2VhHyBg=
+github.com/go-sql-driver/mysql v1.7.0 h1:ueSltNNllEqE3qcWBTD0iQd3IpL/6U+mJxLkazJ7YPc=
+github.com/go-sql-driver/mysql v1.7.0/go.mod h1:OXbVy3sEdcQ2Doequ6Z5BW6fXNQTmx+9S1MCJN5yJMI=
 github.com/goccy/go-json v0.9.7/go.mod h1:6MelG93GURQebXPDq3khkgXZkazVtN9CRI+MGFi0w8I=
 github.com/goccy/go-json v0.10.2 h1:CrxCmQqYDkv1z7lO7Wbh2HN93uovUHgrECaO5ZrCXAU=
 github.com/goccy/go-json v0.10.2/go.mod h1:6MelG93GURQebXPDq3khkgXZkazVtN9CRI+MGFi0w8I=
@@ -90,6 +97,8 @@ github.com/google/go-cmp v0.5.5/go.mod h1:v8dTdLbMG2kIc/vJvl+f65V22dbkXbowE6jgT/
 github.com/google/go-cmp v0.6.0 h1:ofyhxvXcZhMsU5ulbFiLKl/XBFqE1GSq7atu8tAmTRI=
 github.com/google/go-cmp v0.6.0/go.mod h1:17dUlkBOakJ0+DkrSSNjCkIjxS6bF9zb3elmeNGIjoY=
 github.com/google/gofuzz v1.0.0/go.mod h1:dBl0BpW6vV/+mYPU4Po3pmUjxk6FQPldtuIdl/M65Eg=
+github.com/google/pprof v0.0.0-20221118152302-e6195bd50e26 h1:Xim43kblpZXfIBQsbuBVKCudVG457BR2GZFIz3uw3hQ=
+github.com/google/pprof v0.0.0-20221118152302-e6195bd50e26/go.mod h1:dDKJzRmX4S37WGHujM7tX//fmj1uioxKzKxz3lo4HJo=
 github.com/google/uuid v1.6.0 h1:NIvaJDMOsjHA8n1jAhLSgzrAzy1Hgr+hNrb57e+94F0=
 github.com/google/uuid v1.6.0/go.mod h1:TIyPZe4MgqvfeYDBFedMoGGpEw/LqOeaOT+nhxU+yHo=
 github.com/gorilla/context v1.1.1 h1:AWwleXJkX/nhcU9bZSnZoi3h/qGYqQAGhq6zZe/aQW8=
@@ -108,8 +117,6 @@ github.com/jackc/pgx/v5 v5.7.1 h1:x7SYsPBYDkHDksogeSmZZ5xzThcTgRz++I5E+ePFUcs=
 github.com/jackc/pgx/v5 v5.7.1/go.mod h1:e7O26IywZZ+naJtWWos6i6fvWK+29etgITqrqHLfoZA=
 github.com/jackc/puddle/v2 v2.2.2 h1:PR8nw+E/1w0GLuRFSmiioY6UooMp6KJv0/61nB7icHo=
 github.com/jackc/puddle/v2 v2.2.2/go.mod h1:vriiEXHvEE654aYKXXjOvZM39qJ0q+azkZFrfEOc3H4=
-github.com/jinzhu/copier v0.4.0 h1:w3ciUoD19shMCRargcpm0cm91ytaBhDvuRpz1ODO/U8=
-github.com/jinzhu/copier v0.4.0/go.mod h1:DfbEm0FYsaqBcKcFuvmOZb218JkPGtvSHsKg8S8hyyg=
 github.com/jinzhu/inflection v1.0.0 h1:K317FqzuhWc8YvSVlFMCCUb36O/S9MCKRDI7QkRKD/E=
 github.com/jinzhu/inflection v1.0.0/go.mod h1:h+uFLlag+Qp1Va5pdKtLDYj+kHp5pxUVkryuEj+Srlc=
 github.com/jinzhu/now v1.1.4/go.mod h1:d3SSVoowX0Lcu0IBviAWJpolVfI5UJVZZ7cO71lE/z8=
@@ -140,9 +147,6 @@ github.com/mattn/go-isatty v0.0.12/go.mod h1:cbi8OIDigv2wuxKPP5vlRcQ1OAZbq2CE4Ky
 github.com/mattn/go-isatty v0.0.14/go.mod h1:7GGIvUiUoEMVVmxf/4nioHXj79iQHKdU27kJ6hsGG94=
 github.com/mattn/go-isatty v0.0.20 h1:xfD0iDuEKnDkl03q4limB+vH+GxLEtL/jb4xVJSWWEY=
 github.com/mattn/go-isatty v0.0.20/go.mod h1:W+V8PltTTMOvKvAeJH7IuucS94S2C6jfK/D7dTCTo3Y=
-github.com/mattn/go-sqlite3 v1.14.15/go.mod h1:2eHXhiwb8IkHr+BDWZGa96P6+rkvnG63S2DGjv9HUNg=
-github.com/mattn/go-sqlite3 v2.0.3+incompatible h1:gXHsfypPkaMZrKbD5209QV9jbUTJKjyR5WD3HYQSd+U=
-github.com/mattn/go-sqlite3 v2.0.3+incompatible/go.mod h1:FPy6KqzDD04eiIsT53CuJW3U88zkxoIYsOqkbpncsNc=
 github.com/mitchellh/mapstructure v1.5.0 h1:jeMsZIYE/09sWLaz43PL7Gy6RuMjD2eJVyuac5Z2hdY=
 github.com/mitchellh/mapstructure v1.5.0/go.mod h1:bFUtVrKA4DC2yAKiSyO/QUcy7e+RRV2QTWOzhPopBRo=
 github.com/modern-go/concurrent v0.0.0-20180228061459-e0a39a4cb421/go.mod h1:6dJC0mAP4ikYIbvyc7fijjWJddQyLn8Ig3JB5CqoB9Q=
@@ -167,6 +171,9 @@ github.com/pkoukk/tiktoken-go v0.1.7 h1:qOBHXX4PHtvIvmOtyg1EeKlwFRiMKAcoMp4Q+bLQ
 github.com/pkoukk/tiktoken-go v0.1.7/go.mod h1:9NiV+i9mJKGj1rYOT+njbv+ZwA/zJxYdewGl6qVatpg=
 github.com/pmezard/go-difflib v1.0.0 h1:4DBwDE0NGyQoBHbLQYPwSUPoCMWR5BEzIk/f1lZbAQM=
 github.com/pmezard/go-difflib v1.0.0/go.mod h1:iKH77koFhYxTK1pcRnkKkqfTogsbg7gZNVY4sRDYZ/4=
+github.com/remyoudompheng/bigfft v0.0.0-20200410134404-eec4a21b6bb0/go.mod h1:qqbHyh8v60DhA7CoWK5oRCqLrMHRGoxYCSS9EjAz6Eo=
+github.com/remyoudompheng/bigfft v0.0.0-20230129092748-24d4a6f8daec h1:W09IVJc94icq4NjY3clb7Lk8O1qJ8BdBEF8z0ibU0rE=
+github.com/remyoudompheng/bigfft v0.0.0-20230129092748-24d4a6f8daec/go.mod h1:qqbHyh8v60DhA7CoWK5oRCqLrMHRGoxYCSS9EjAz6Eo=
 github.com/rogpeppe/go-internal v1.6.1/go.mod h1:xXDCJY+GAPziupqXw64V24skbSoqbTEfhy4qGm1nDQc=
 github.com/rogpeppe/go-internal v1.8.0 h1:FCbCCtXNOY3UtUuHUYaghJg4y7Fd14rXifAYUAtL9R8=
 github.com/rogpeppe/go-internal v1.8.0/go.mod h1:WmiCO8CzOY8rg0OYDC4/i/2WRWAB6poM+XZ2dLUbcbE=
@@ -174,6 +181,8 @@ github.com/samber/lo v1.39.0 h1:4gTz1wUhNYLhFSKl6O+8peW0v2F4BCY034GRpU9WnuA=
 github.com/samber/lo v1.39.0/go.mod h1:+m/ZKRl6ClXCE2Lgf3MsQlWfh4bn1bz6CXEOxnEXnEA=
 github.com/shirou/gopsutil v3.21.11+incompatible h1:+1+c1VGhc88SSonWP6foOcLhvnKlUeu/erjjvaPEYiI=
 github.com/shirou/gopsutil v3.21.11+incompatible/go.mod h1:5b4v6he4MtMOwMlS0TUMTu2PcXUg8+E1lC7eC3UO/RA=
+github.com/shopspring/decimal v1.4.0 h1:bxl37RwXBklmTi0C79JfXCEBD1cqqHt0bbgBAGFp81k=
+github.com/shopspring/decimal v1.4.0/go.mod h1:gawqmDU56v4yIKSwfBSFip1HdCCXN8/+DMd9qYNcwME=
 github.com/stretchr/objx v0.1.0/go.mod h1:HFkY916IF+rwdDfMAkV7OtwuqBVzrE8GR6GFx+wExME=
 github.com/stretchr/objx v0.4.0/go.mod h1:YvHI0jy2hoMjB+UWwv71VJQ9isScKT/TqJzVSSt89Yw=
 github.com/stretchr/objx v0.5.0/go.mod h1:Yh+to48EsGEfYuaHDzXPcE3xhTkx73EhmCGUpEOglKo=
@@ -263,11 +272,16 @@ gorm.io/driver/mysql v1.4.3 h1:/JhWJhO2v17d8hjApTltKNADm7K7YI2ogkR7avJUL3k=
 gorm.io/driver/mysql v1.4.3/go.mod h1:sSIebwZAVPiT+27jK9HIwvsqOGKx3YMPmrA3mBJR10c=
 gorm.io/driver/postgres v1.5.2 h1:ytTDxxEv+MplXOfFe3Lzm7SjG09fcdb3Z/c056DTBx0=
 gorm.io/driver/postgres v1.5.2/go.mod h1:fmpX0m2I1PKuR7mKZiEluwrP3hbs+ps7JIGMUBpCgl8=
-gorm.io/driver/sqlite v1.4.3 h1:HBBcZSDnWi5BW3B3rwvVTc510KGkBkexlOg0QrmLUuU=
-gorm.io/driver/sqlite v1.4.3/go.mod h1:0Aq3iPO+v9ZKbcdiz8gLWRw5VOPcBOPUQJFLq5e2ecI=
 gorm.io/gorm v1.23.8/go.mod h1:l2lP/RyAtc1ynaTjFksBde/O8v9oOGIApu2/xRitmZk=
-gorm.io/gorm v1.24.0/go.mod h1:DVrVomtaYTbqs7gB/x2uVvqnXzv0nqjB396B8cG4dBA=
-gorm.io/gorm v1.25.0 h1:+KtYtb2roDz14EQe4bla8CbQlmb9dN3VejSai3lprfU=
-gorm.io/gorm v1.25.0/go.mod h1:L4uxeKpfBml98NYqVqwAdmV1a2nBtAec/cf3fpucW/k=
+gorm.io/gorm v1.25.2 h1:gs1o6Vsa+oVKG/a9ElL3XgyGfghFfkKA2SInQaCyMho=
+gorm.io/gorm v1.25.2/go.mod h1:L4uxeKpfBml98NYqVqwAdmV1a2nBtAec/cf3fpucW/k=
+modernc.org/libc v1.22.5 h1:91BNch/e5B0uPbJFgqbxXuOnxBQjlS//icfQEGmvyjE=
+modernc.org/libc v1.22.5/go.mod h1:jj+Z7dTNX8fBScMVNRAYZ/jF91K8fdT2hYMThc3YjBY=
+modernc.org/mathutil v1.5.0 h1:rV0Ko/6SfM+8G+yKiyI830l3Wuz1zRutdslNoQ0kfiQ=
+modernc.org/mathutil v1.5.0/go.mod h1:mZW8CKdRPY1v87qxC/wUdX5O1qDzXMP5TH3wjfpga6E=
+modernc.org/memory v1.5.0 h1:N+/8c5rE6EqugZwHii4IFsaJ7MUhoWX07J5tC/iI5Ds=
+modernc.org/memory v1.5.0/go.mod h1:PkUhL0Mugw21sHPeskwZW4D6VscE/GQJOnIpCnW6pSU=
+modernc.org/sqlite v1.23.1 h1:nrSBg4aRQQwq59JpvGEQ15tNxoO5pX/kUjcRNwSAGQM=
+modernc.org/sqlite v1.23.1/go.mod h1:OrDj17Mggn6MhE+iPbBNf7RGKODDE9NFT0f3EwDzJqk=
 nullprogram.com/x/optparse v1.0.0/go.mod h1:KdyPE+Igbe0jQUrVfMqDMeJQIJZEuyV7pjYmp6pbG50=
 rsc.io/pdf v0.1.1/go.mod h1:n8OzWcQ6Sp37PL01nO98y4iUCRdTGarVfzxY20ICaU4=

main.go

@@ -119,9 +119,9 @@ func main() {
 	}
 
 	if os.Getenv("ENABLE_PPROF") == "true" {
-		go func() {
+		gopool.Go(func() {
 			log.Println(http.ListenAndServe("0.0.0.0:8005", nil))
-		}()
+		})
 		go common.Monitor()
 		common.SysLog("pprof enabled")
 	}
@@ -145,6 +145,13 @@ func main() {
 	middleware.SetUpLogger(server)
 	// Initialize session store
 	store := cookie.NewStore([]byte(common.SessionSecret))
+	store.Options(sessions.Options{
+		Path:     "/",
+		MaxAge:   2592000, // 30 days
+		HttpOnly: true,
+		Secure:   false,
+		SameSite: http.SameSiteStrictMode,
+	})
 	server.Use(sessions.Sessions("session", store))
 
 	router.SetRouter(server, buildFS, indexPage)

middleware/auth.go

@@ -199,15 +199,19 @@ func TokenAuth() func(c *gin.Context) {
 			abortWithOpenAiMessage(c, http.StatusUnauthorized, err.Error())
 			return
 		}
-		userEnabled, err := model.IsUserEnabled(token.UserId, false)
+		userCache, err := model.GetUserCache(token.UserId)
 		if err != nil {
 			abortWithOpenAiMessage(c, http.StatusInternalServerError, err.Error())
 			return
 		}
+		userEnabled := userCache.Status == common.UserStatusEnabled
 		if !userEnabled {
 			abortWithOpenAiMessage(c, http.StatusForbidden, "用户已被封禁")
 			return
 		}
+
+		userCache.WriteContext(c)
+
 		c.Set("id", token.UserId)
 		c.Set("token_id", token.Id)
 		c.Set("token_key", token.Key)

middleware/distributor.go

@@ -32,7 +32,6 @@ func Distribute() func(c *gin.Context) {
 				return
 			}
 		}
-		userId := c.GetInt("id")
 		var channel *model.Channel
 		channelId, ok := c.Get("specific_channel_id")
 		modelRequest, shouldSelectChannel, err := getModelRequest(c)
@@ -40,7 +39,7 @@ func Distribute() func(c *gin.Context) {
 			abortWithOpenAiMessage(c, http.StatusBadRequest, "Invalid request, "+err.Error())
 			return
 		}
-		userGroup, _ := model.GetUserGroup(userId, false)
+		userGroup := c.GetString(constant.ContextKeyUserGroup)
 		tokenGroup := c.GetString("token_group")
 		if tokenGroup != "" {
 			// check common.UserUsableGroups[userGroup]
@@ -135,17 +134,14 @@ func getModelRequest(c *gin.Context) (*ModelRequest, bool, error) {
 			midjourneyRequest := dto.MidjourneyRequest{}
 			err = common.UnmarshalBodyReusable(c, &midjourneyRequest)
 			if err != nil {
-				abortWithMidjourneyMessage(c, http.StatusBadRequest, constant.MjErrorUnknown, "无效的请求, "+err.Error())
 				return nil, false, err
 			}
 			midjourneyModel, mjErr, success := service.GetMjRequestModel(relayMode, &midjourneyRequest)
 			if mjErr != nil {
-				abortWithMidjourneyMessage(c, http.StatusBadRequest, mjErr.Code, mjErr.Description)
 				return nil, false, fmt.Errorf(mjErr.Description)
 			}
 			if midjourneyModel == "" {
 				if !success {
-					abortWithMidjourneyMessage(c, http.StatusBadRequest, constant.MjErrorUnknown, "无效的请求, 无法解析模型")
 					return nil, false, fmt.Errorf("无效的请求, 无法解析模型")
 				} else {
 					// task fetch, task fetch by condition, notify
@@ -170,7 +166,6 @@ func getModelRequest(c *gin.Context) (*ModelRequest, bool, error) {
 		err = common.UnmarshalBodyReusable(c, &modelRequest)
 	}
 	if err != nil {
-		abortWithOpenAiMessage(c, http.StatusBadRequest, "无效的请求, "+err.Error())
 		return nil, false, errors.New("无效的请求, " + err.Error())
 	}
 	if strings.HasPrefix(c.Request.URL.Path, "/v1/realtime") {
@@ -239,5 +234,7 @@ func SetupContextForSelectedChannel(c *gin.Context, channel *model.Channel, mode
 		c.Set("plugin", channel.Other)
 	case common.ChannelCloudflare:
 		c.Set("api_version", channel.Other)
+	case common.ChannelTypeMokaAI:
+		c.Set("api_version", channel.Other)
 	}
 }

middleware/model-rate-limit.go

@@ -0,0 +1,175 @@
+package middleware
+
+import (
+	"context"
+	"fmt"
+	"net/http"
+	"one-api/common"
+	"one-api/setting"
+	"strconv"
+	"time"
+
+	"github.com/gin-gonic/gin"
+	"github.com/go-redis/redis/v8"
+)
+
+const (
+	ModelRequestRateLimitCountMark        = "MRRL"
+	ModelRequestRateLimitSuccessCountMark = "MRRLS"
+)
+
+// 检查Redis中的请求限制
+func checkRedisRateLimit(ctx context.Context, rdb *redis.Client, key string, maxCount int, duration int64) (bool, error) {
+	// 如果maxCount为0，表示不限制
+	if maxCount == 0 {
+		return true, nil
+	}
+
+	// 获取当前计数
+	length, err := rdb.LLen(ctx, key).Result()
+	if err != nil {
+		return false, err
+	}
+
+	// 如果未达到限制，允许请求
+	if length < int64(maxCount) {
+		return true, nil
+	}
+
+	// 检查时间窗口
+	oldTimeStr, _ := rdb.LIndex(ctx, key, -1).Result()
+	oldTime, err := time.Parse(timeFormat, oldTimeStr)
+	if err != nil {
+		return false, err
+	}
+
+	nowTimeStr := time.Now().Format(timeFormat)
+	nowTime, err := time.Parse(timeFormat, nowTimeStr)
+	if err != nil {
+		return false, err
+	}
+	// 如果在时间窗口内已达到限制，拒绝请求
+	subTime := nowTime.Sub(oldTime).Seconds()
+	if int64(subTime) < duration {
+		rdb.Expire(ctx, key, time.Duration(setting.ModelRequestRateLimitDurationMinutes)*time.Minute)
+		return false, nil
+	}
+
+	return true, nil
+}
+
+// 记录Redis请求
+func recordRedisRequest(ctx context.Context, rdb *redis.Client, key string, maxCount int) {
+	// 如果maxCount为0，不记录请求
+	if maxCount == 0 {
+		return
+	}
+
+	now := time.Now().Format(timeFormat)
+	rdb.LPush(ctx, key, now)
+	rdb.LTrim(ctx, key, 0, int64(maxCount-1))
+	rdb.Expire(ctx, key, time.Duration(setting.ModelRequestRateLimitDurationMinutes)*time.Minute)
+}
+
+// Redis限流处理器
+func redisRateLimitHandler(duration int64, totalMaxCount, successMaxCount int) gin.HandlerFunc {
+	return func(c *gin.Context) {
+		userId := strconv.Itoa(c.GetInt("id"))
+		ctx := context.Background()
+		rdb := common.RDB
+
+		// 1. 检查总请求数限制（当totalMaxCount为0时会自动跳过）
+		totalKey := fmt.Sprintf("rateLimit:%s:%s", ModelRequestRateLimitCountMark, userId)
+		allowed, err := checkRedisRateLimit(ctx, rdb, totalKey, totalMaxCount, duration)
+		if err != nil {
+			fmt.Println("检查总请求数限制失败:", err.Error())
+			abortWithOpenAiMessage(c, http.StatusInternalServerError, "rate_limit_check_failed")
+			return
+		}
+		if !allowed {
+			abortWithOpenAiMessage(c, http.StatusTooManyRequests, fmt.Sprintf("您已达到总请求数限制：%d分钟内最多请求%d次，包括失败次数，请检查您的请求是否正确", setting.ModelRequestRateLimitDurationMinutes, totalMaxCount))
+		}
+
+		// 2. 检查成功请求数限制
+		successKey := fmt.Sprintf("rateLimit:%s:%s", ModelRequestRateLimitSuccessCountMark, userId)
+		allowed, err = checkRedisRateLimit(ctx, rdb, successKey, successMaxCount, duration)
+		if err != nil {
+			fmt.Println("检查成功请求数限制失败:", err.Error())
+			abortWithOpenAiMessage(c, http.StatusInternalServerError, "rate_limit_check_failed")
+			return
+		}
+		if !allowed {
+			abortWithOpenAiMessage(c, http.StatusTooManyRequests, fmt.Sprintf("您已达到请求数限制：%d分钟内最多请求%d次", setting.ModelRequestRateLimitDurationMinutes, successMaxCount))
+			return
+		}
+
+		// 3. 记录总请求（当totalMaxCount为0时会自动跳过）
+		recordRedisRequest(ctx, rdb, totalKey, totalMaxCount)
+
+		// 4. 处理请求
+		c.Next()
+
+		// 5. 如果请求成功，记录成功请求
+		if c.Writer.Status() < 400 {
+			recordRedisRequest(ctx, rdb, successKey, successMaxCount)
+		}
+	}
+}
+
+// 内存限流处理器
+func memoryRateLimitHandler(duration int64, totalMaxCount, successMaxCount int) gin.HandlerFunc {
+	inMemoryRateLimiter.Init(time.Duration(setting.ModelRequestRateLimitDurationMinutes) * time.Minute)
+
+	return func(c *gin.Context) {
+		userId := strconv.Itoa(c.GetInt("id"))
+		totalKey := ModelRequestRateLimitCountMark + userId
+		successKey := ModelRequestRateLimitSuccessCountMark + userId
+
+		// 1. 检查总请求数限制（当totalMaxCount为0时跳过）
+		if totalMaxCount > 0 && !inMemoryRateLimiter.Request(totalKey, totalMaxCount, duration) {
+			c.Status(http.StatusTooManyRequests)
+			c.Abort()
+			return
+		}
+
+		// 2. 检查成功请求数限制
+		// 使用一个临时key来检查限制，这样可以避免实际记录
+		checkKey := successKey + "_check"
+		if !inMemoryRateLimiter.Request(checkKey, successMaxCount, duration) {
+			c.Status(http.StatusTooManyRequests)
+			c.Abort()
+			return
+		}
+
+		// 3. 处理请求
+		c.Next()
+
+		// 4. 如果请求成功，记录到实际的成功请求计数中
+		if c.Writer.Status() < 400 {
+			inMemoryRateLimiter.Request(successKey, successMaxCount, duration)
+		}
+	}
+}
+
+// ModelRequestRateLimit 模型请求限流中间件
+func ModelRequestRateLimit() func(c *gin.Context) {
+	return func(c *gin.Context) {
+		// 在每个请求时检查是否启用限流
+		if !setting.ModelRequestRateLimitEnabled {
+			c.Next()
+			return
+		}
+
+		// 计算限流参数
+		duration := int64(setting.ModelRequestRateLimitDurationMinutes * 60)
+		totalMaxCount := setting.ModelRequestRateLimitCount
+		successMaxCount := setting.ModelRequestRateLimitSuccessCount
+
+		// 根据存储类型选择并执行限流处理器
+		if common.RedisEnabled {
+			redisRateLimitHandler(duration, totalMaxCount, successMaxCount)(c)
+		} else {
+			memoryRateLimitHandler(duration, totalMaxCount, successMaxCount)(c)
+		}
+	}
+}

model/cache.go

@@ -11,106 +11,6 @@ import (
 	"time"
 )
 
-//func CacheGetUserGroup(id int) (group string, err error) {
-//	if !common.RedisEnabled {
-//		return GetUserGroup(id)
-//	}
-//	group, err = common.RedisGet(fmt.Sprintf("user_group:%d", id))
-//	if err != nil {
-//		group, err = GetUserGroup(id)
-//		if err != nil {
-//			return "", err
-//		}
-//		err = common.RedisSet(fmt.Sprintf("user_group:%d", id), group, time.Duration(constant.UserId2GroupCacheSeconds)*time.Second)
-//		if err != nil {
-//			common.SysError("Redis set user group error: " + err.Error())
-//		}
-//	}
-//	return group, err
-//}
-//
-//func CacheGetUsername(id int) (username string, err error) {
-//	if !common.RedisEnabled {
-//		return GetUsernameById(id)
-//	}
-//	username, err = common.RedisGet(fmt.Sprintf("user_name:%d", id))
-//	if err != nil {
-//		username, err = GetUsernameById(id)
-//		if err != nil {
-//			return "", err
-//		}
-//		err = common.RedisSet(fmt.Sprintf("user_name:%d", id), username, time.Duration(constant.UserId2GroupCacheSeconds)*time.Second)
-//		if err != nil {
-//			common.SysError("Redis set user group error: " + err.Error())
-//		}
-//	}
-//	return username, err
-//}
-//
-//func CacheGetUserQuota(id int) (quota int, err error) {
-//	if !common.RedisEnabled {
-//		return GetUserQuota(id)
-//	}
-//	quotaString, err := common.RedisGet(fmt.Sprintf("user_quota:%d", id))
-//	if err != nil {
-//		quota, err = GetUserQuota(id)
-//		if err != nil {
-//			return 0, err
-//		}
-//		return quota, nil
-//	}
-//	quota, err = strconv.Atoi(quotaString)
-//	return quota, nil
-//}
-//
-//func CacheUpdateUserQuota(id int) error {
-//	if !common.RedisEnabled {
-//		return nil
-//	}
-//	quota, err := GetUserQuota(id)
-//	if err != nil {
-//		return err
-//	}
-//	return cacheSetUserQuota(id, quota)
-//}
-//
-//func cacheSetUserQuota(id int, quota int) error {
-//	err := common.RedisSet(fmt.Sprintf("user_quota:%d", id), fmt.Sprintf("%d", quota), time.Duration(constant.UserId2QuotaCacheSeconds)*time.Second)
-//	return err
-//}
-//
-//func CacheDecreaseUserQuota(id int, quota int) error {
-//	if !common.RedisEnabled {
-//		return nil
-//	}
-//	err := common.RedisDecrease(fmt.Sprintf("user_quota:%d", id), int64(quota))
-//	return err
-//}
-//
-//func CacheIsUserEnabled(userId int) (bool, error) {
-//	if !common.RedisEnabled {
-//		return IsUserEnabled(userId)
-//	}
-//	enabled, err := common.RedisGet(fmt.Sprintf("user_enabled:%d", userId))
-//	if err == nil {
-//		return enabled == "1", nil
-//	}
-//
-//	userEnabled, err := IsUserEnabled(userId)
-//	if err != nil {
-//		return false, err
-//	}
-//	enabled = "0"
-//	if userEnabled {
-//		enabled = "1"
-//	}
-//	err = common.RedisSet(fmt.Sprintf("user_enabled:%d", userId), enabled, time.Duration(constant.UserId2StatusCacheSeconds)*time.Second)
-//	if err != nil {
-//		common.SysError("Redis set user enabled error: " + err.Error())
-//	}
-//	return userEnabled, err
-//}
-
 var group2model2channels map[string]map[string][]*Channel
 var channelsIDM map[int]*Channel
 var channelSyncLock sync.RWMutex

model/channel.go

@@ -28,14 +28,14 @@ type Channel struct {
 	Models             string  `json:"models"`
 	Group              string  `json:"group" gorm:"type:varchar(64);default:'default'"`
 	UsedQuota          int64   `json:"used_quota" gorm:"bigint;default:0"`
-	ModelMapping       *string `json:"model_mapping" gorm:"type:varchar(1024);default:''"`
+	ModelMapping       *string `json:"model_mapping" gorm:"type:text"`
 	//MaxInputTokens     *int    `json:"max_input_tokens" gorm:"default:0"`
 	StatusCodeMapping *string `json:"status_code_mapping" gorm:"type:varchar(1024);default:''"`
 	Priority          *int64  `json:"priority" gorm:"bigint;default:0"`
 	AutoBan           *int    `json:"auto_ban" gorm:"default:1"`
 	OtherInfo         string  `json:"other_info"`
 	Tag               *string `json:"tag" gorm:"index"`
-	Setting           string  `json:"setting" gorm:"type:text"`
+	Setting           *string `json:"setting" gorm:"type:text"`
 }
 
 func (channel *Channel) GetModels() []string {
@@ -290,35 +290,42 @@ func (channel *Channel) Delete() error {
 
 var channelStatusLock sync.Mutex
 
-func UpdateChannelStatusById(id int, status int, reason string) {
+func UpdateChannelStatusById(id int, status int, reason string) bool {
 	if common.MemoryCacheEnabled {
 		channelStatusLock.Lock()
+		defer channelStatusLock.Unlock()
+
 		channelCache, _ := CacheGetChannel(id)
 		// 如果缓存渠道存在，且状态已是目标状态，直接返回
 		if channelCache != nil && channelCache.Status == status {
-			channelStatusLock.Unlock()
-			return
+			return false
 		}
 		// 如果缓存渠道不存在(说明已经被禁用)，且要设置的状态不为启用，直接返回
 		if channelCache == nil && status != common.ChannelStatusEnabled {
-			channelStatusLock.Unlock()
-			return
+			return false
 		}
 		CacheUpdateChannelStatus(id, status)
-		channelStatusLock.Unlock()
 	}
 	err := UpdateAbilityStatus(id, status == common.ChannelStatusEnabled)
 	if err != nil {
 		common.SysError("failed to update ability status: " + err.Error())
+		return false
 	}
 	channel, err := GetChannelById(id, true)
 	if err != nil {
 		// find channel by id error, directly update status
-		err = DB.Model(&Channel{}).Where("id = ?", id).Update("status", status).Error
-		if err != nil {
-			common.SysError("failed to update channel status: " + err.Error())
+		result := DB.Model(&Channel{}).Where("id = ?", id).Update("status", status)
+		if result.Error != nil {
+			common.SysError("failed to update channel status: " + result.Error.Error())
+			return false
+		}
+		if result.RowsAffected == 0 {
+			return false
 		}
 	} else {
+		if channel.Status == status {
+			return false
+		}
 		// find channel by id success, update status and other info
 		info := channel.GetOtherInfo()
 		info["status_reason"] = reason
@@ -328,9 +335,10 @@ func UpdateChannelStatusById(id int, status int, reason string) {
 		err = channel.Save()
 		if err != nil {
 			common.SysError("failed to update channel status: " + err.Error())
+			return false
 		}
 	}
-
+	return true
 }
 
 func EnableChannelByTag(tag string) error {
@@ -485,8 +493,8 @@ func SearchTags(keyword string, group string, model string, idSort bool) ([]*str
 
 func (channel *Channel) GetSetting() map[string]interface{} {
 	setting := make(map[string]interface{})
-	if channel.Setting != "" {
-		err := json.Unmarshal([]byte(channel.Setting), &setting)
+	if channel.Setting != nil && *channel.Setting != "" {
+		err := json.Unmarshal([]byte(*channel.Setting), &setting)
 		if err != nil {
 			common.SysError("failed to unmarshal setting: " + err.Error())
 		}
@@ -500,7 +508,7 @@ func (channel *Channel) SetSetting(setting map[string]interface{}) {
 		common.SysError("failed to marshal setting: " + err.Error())
 		return
 	}
-	channel.Setting = string(settingBytes)
+	channel.Setting = common.GetPointer[string](string(settingBytes))
 }
 
 func GetChannelsByIds(ids []int) ([]*Channel, error) {

model/log.go

@@ -1,13 +1,14 @@
 package model
 
 import (
-	"context"
 	"fmt"
 	"one-api/common"
 	"os"
 	"strings"
 	"time"
 
+	"github.com/gin-gonic/gin"
+
 	"github.com/bytedance/gopkg/util/gopool"
 	"gorm.io/gorm"
 )
@@ -18,7 +19,7 @@ type Log struct {
 	CreatedAt        int64  `json:"created_at" gorm:"bigint;index:idx_created_at_id,priority:2;index:idx_created_at_type"`
 	Type             int    `json:"type" gorm:"index:idx_created_at_type"`
 	Content          string `json:"content"`
-	Username         string `json:"username" gorm:"index:index_username_model_name,priority:2;default:''"`
+	Username         string `json:"username" gorm:"index;index:index_username_model_name,priority:2;default:''"`
 	TokenName        string `json:"token_name" gorm:"index;default:''"`
 	ModelName        string `json:"model_name" gorm:"index;index:index_username_model_name,priority:1;default:''"`
 	Quota            int    `json:"quota" gorm:"default:0"`
@@ -87,14 +88,14 @@ func RecordLog(userId int, logType int, content string) {
 	}
 }
 
-func RecordConsumeLog(ctx context.Context, userId int, channelId int, promptTokens int, completionTokens int,
+func RecordConsumeLog(c *gin.Context, userId int, channelId int, promptTokens int, completionTokens int,
 	modelName string, tokenName string, quota int, content string, tokenId int, userQuota int, useTimeSeconds int,
 	isStream bool, group string, other map[string]interface{}) {
-	common.LogInfo(ctx, fmt.Sprintf("record consume log: userId=%d, 用户调用前余额=%d, channelId=%d, promptTokens=%d, completionTokens=%d, modelName=%s, tokenName=%s, quota=%d, content=%s", userId, userQuota, channelId, promptTokens, completionTokens, modelName, tokenName, quota, content))
+	common.LogInfo(c, fmt.Sprintf("record consume log: userId=%d, 用户调用前余额=%d, channelId=%d, promptTokens=%d, completionTokens=%d, modelName=%s, tokenName=%s, quota=%d, content=%s", userId, userQuota, channelId, promptTokens, completionTokens, modelName, tokenName, quota, content))
 	if !common.LogConsumeEnabled {
 		return
 	}
-	username, _ := GetUsernameById(userId, false)
+	username := c.GetString("username")
 	otherStr := common.MapToJsonStr(other)
 	log := &Log{
 		UserId:           userId,
@@ -116,7 +117,7 @@ func RecordConsumeLog(ctx context.Context, userId int, channelId int, promptToke
 	}
 	err := LOG_DB.Create(log).Error
 	if err != nil {
-		common.LogError(ctx, "failed to record log: "+err.Error())
+		common.LogError(c, "failed to record log: "+err.Error())
 	}
 	if common.DataExportEnabled {
 		gopool.Go(func() {
@@ -133,9 +134,6 @@ func GetAllLogs(logType int, startTimestamp int64, endTimestamp int64, modelName
 		tx = LOG_DB.Where("logs.type = ?", logType)
 	}
 
-	tx = tx.Joins("LEFT JOIN channels ON logs.channel_id = channels.id")
-	tx = tx.Select("logs.*, channels.name as channel_name")
-
 	if modelName != "" {
 		tx = tx.Where("logs.model_name like ?", modelName)
 	}
@@ -165,6 +163,30 @@ func GetAllLogs(logType int, startTimestamp int64, endTimestamp int64, modelName
 	if err != nil {
 		return nil, 0, err
 	}
+
+	channelIds := make([]int, 0)
+	channelMap := make(map[int]string)
+	for _, log := range logs {
+		if log.ChannelId != 0 {
+			channelIds = append(channelIds, log.ChannelId)
+		}
+	}
+	if len(channelIds) > 0 {
+		var channels []struct {
+			Id   int    `gorm:"column:id"`
+			Name string `gorm:"column:name"`
+		}
+		if err = DB.Table("channels").Select("id, name").Where("id IN ?", channelIds).Find(&channels).Error; err != nil {
+			return logs, total, err
+		}
+		for _, channel := range channels {
+			channelMap[channel.Id] = channel.Name
+		}
+		for i := range logs {
+			logs[i].ChannelName = channelMap[logs[i].ChannelId]
+		}
+	}
+
 	return logs, total, err
 }
 
@@ -176,9 +198,6 @@ func GetUserLogs(userId int, logType int, startTimestamp int64, endTimestamp int
 		tx = LOG_DB.Where("logs.user_id = ? and logs.type = ?", userId, logType)
 	}
 
-	tx = tx.Joins("LEFT JOIN channels ON logs.channel_id = channels.id")
-	tx = tx.Select("logs.*, channels.name as channel_name")
-
 	if modelName != "" {
 		tx = tx.Where("logs.model_name like ?", modelName)
 	}
@@ -199,6 +218,10 @@ func GetUserLogs(userId int, logType int, startTimestamp int64, endTimestamp int
 		return nil, 0, err
 	}
 	err = tx.Order("logs.id desc").Limit(num).Offset(startIdx).Find(&logs).Error
+	if err != nil {
+		return nil, 0, err
+	}
+
 	formatUserLogs(logs)
 	return logs, total, err
 }

model/main.go

@@ -1,9 +1,9 @@
 package model
 
 import (
+	"github.com/glebarez/sqlite"
 	"gorm.io/driver/mysql"
 	"gorm.io/driver/postgres"
-	"gorm.io/driver/sqlite"
 	"gorm.io/gorm"
 	"log"
 	"one-api/common"
@@ -119,12 +119,9 @@ func InitDB() (err error) {
 		if !common.IsMasterNode {
 			return nil
 		}
-		//if common.UsingMySQL {
-		//	_, _ = sqlDB.Exec("DROP INDEX idx_channels_key ON channels;")             // TODO: delete this line when most users have upgraded
-		//	_, _ = sqlDB.Exec("ALTER TABLE midjourneys MODIFY action VARCHAR(40);")   // TODO: delete this line when most users have upgraded
-		//	_, _ = sqlDB.Exec("ALTER TABLE midjourneys MODIFY progress VARCHAR(30);") // TODO: delete this line when most users have upgraded
-		//	_, _ = sqlDB.Exec("ALTER TABLE midjourneys MODIFY status VARCHAR(20);")   // TODO: delete this line when most users have upgraded
-		//}
+		if common.UsingMySQL {
+			_, _ = sqlDB.Exec("ALTER TABLE channels MODIFY model_mapping TEXT;") // TODO: delete this line when most users have upgraded
+		}
 		common.SysLog("database migration started")
 		err = migrateDB()
 		return err

model/option.go

@@ -3,6 +3,8 @@ package model
 import (
 	"one-api/common"
 	"one-api/setting"
+	"one-api/setting/config"
+	"one-api/setting/operation_setting"
 	"strconv"
 	"strings"
 	"time"
@@ -23,6 +25,8 @@ func AllOption() ([]*Option, error) {
 func InitOptionMap() {
 	common.OptionMapRWMutex.Lock()
 	common.OptionMap = make(map[string]string)
+
+	// 添加原有的系统配置
 	common.OptionMap["FileUploadPermission"] = strconv.Itoa(common.FileUploadPermission)
 	common.OptionMap["FileDownloadPermission"] = strconv.Itoa(common.FileDownloadPermission)
 	common.OptionMap["ImageUploadPermission"] = strconv.Itoa(common.ImageUploadPermission)
@@ -85,14 +89,18 @@ func InitOptionMap() {
 	common.OptionMap["QuotaForInvitee"] = strconv.Itoa(common.QuotaForInvitee)
 	common.OptionMap["QuotaRemindThreshold"] = strconv.Itoa(common.QuotaRemindThreshold)
 	common.OptionMap["PreConsumedQuota"] = strconv.Itoa(common.PreConsumedQuota)
-	common.OptionMap["ModelRatio"] = common.ModelRatio2JSONString()
-	common.OptionMap["ModelPrice"] = common.ModelPrice2JSONString()
+	common.OptionMap["ModelRequestRateLimitCount"] = strconv.Itoa(setting.ModelRequestRateLimitCount)
+	common.OptionMap["ModelRequestRateLimitDurationMinutes"] = strconv.Itoa(setting.ModelRequestRateLimitDurationMinutes)
+	common.OptionMap["ModelRequestRateLimitSuccessCount"] = strconv.Itoa(setting.ModelRequestRateLimitSuccessCount)
+	common.OptionMap["ModelRatio"] = operation_setting.ModelRatio2JSONString()
+	common.OptionMap["ModelPrice"] = operation_setting.ModelPrice2JSONString()
+	common.OptionMap["CacheRatio"] = operation_setting.CacheRatio2JSONString()
 	common.OptionMap["GroupRatio"] = setting.GroupRatio2JSONString()
 	common.OptionMap["UserUsableGroups"] = setting.UserUsableGroups2JSONString()
-	common.OptionMap["CompletionRatio"] = common.CompletionRatio2JSONString()
+	common.OptionMap["CompletionRatio"] = operation_setting.CompletionRatio2JSONString()
 	common.OptionMap["TopUpLink"] = common.TopUpLink
-	common.OptionMap["ChatLink"] = common.ChatLink
-	common.OptionMap["ChatLink2"] = common.ChatLink2
+	//common.OptionMap["ChatLink"] = common.ChatLink
+	//common.OptionMap["ChatLink2"] = common.ChatLink2
 	common.OptionMap["QuotaPerUnit"] = strconv.FormatFloat(common.QuotaPerUnit, 'f', -1, 64)
 	common.OptionMap["RetryTimes"] = strconv.Itoa(common.RetryTimes)
 	common.OptionMap["DataExportInterval"] = strconv.Itoa(common.DataExportInterval)
@@ -104,11 +112,20 @@ func InitOptionMap() {
 	common.OptionMap["MjForwardUrlEnabled"] = strconv.FormatBool(setting.MjForwardUrlEnabled)
 	common.OptionMap["MjActionCheckSuccessEnabled"] = strconv.FormatBool(setting.MjActionCheckSuccessEnabled)
 	common.OptionMap["CheckSensitiveEnabled"] = strconv.FormatBool(setting.CheckSensitiveEnabled)
+	common.OptionMap["DemoSiteEnabled"] = strconv.FormatBool(operation_setting.DemoSiteEnabled)
+	common.OptionMap["SelfUseModeEnabled"] = strconv.FormatBool(operation_setting.SelfUseModeEnabled)
+	common.OptionMap["ModelRequestRateLimitEnabled"] = strconv.FormatBool(setting.ModelRequestRateLimitEnabled)
 	common.OptionMap["CheckSensitiveOnPromptEnabled"] = strconv.FormatBool(setting.CheckSensitiveOnPromptEnabled)
-	//common.OptionMap["CheckSensitiveOnCompletionEnabled"] = strconv.FormatBool(constant.CheckSensitiveOnCompletionEnabled)
 	common.OptionMap["StopOnSensitiveEnabled"] = strconv.FormatBool(setting.StopOnSensitiveEnabled)
 	common.OptionMap["SensitiveWords"] = setting.SensitiveWordsToString()
 	common.OptionMap["StreamCacheQueueLength"] = strconv.Itoa(setting.StreamCacheQueueLength)
+	common.OptionMap["AutomaticDisableKeywords"] = operation_setting.AutomaticDisableKeywordsToString()
+
+	// 自动添加所有注册的模型配置
+	modelConfigs := config.GlobalConfig.ExportAllConfigs()
+	for k, v := range modelConfigs {
+		common.OptionMap[k] = v
+	}
 
 	common.OptionMapRWMutex.Unlock()
 	loadOptionsFromDatabase()
@@ -152,6 +169,13 @@ func updateOptionMap(key string, value string) (err error) {
 	common.OptionMapRWMutex.Lock()
 	defer common.OptionMapRWMutex.Unlock()
 	common.OptionMap[key] = value
+
+	// 检查是否是模型配置 - 使用更规范的方式处理
+	if handleConfigUpdate(key, value) {
+		return nil // 已由配置系统处理
+	}
+
+	// 处理传统配置项...
 	if strings.HasSuffix(key, "Permission") {
 		intValue, _ := strconv.Atoi(value)
 		switch key {
@@ -220,10 +244,14 @@ func updateOptionMap(key string, value string) (err error) {
 			setting.MjActionCheckSuccessEnabled = boolValue
 		case "CheckSensitiveEnabled":
 			setting.CheckSensitiveEnabled = boolValue
+		case "DemoSiteEnabled":
+			operation_setting.DemoSiteEnabled = boolValue
+		case "SelfUseModeEnabled":
+			operation_setting.SelfUseModeEnabled = boolValue
 		case "CheckSensitiveOnPromptEnabled":
 			setting.CheckSensitiveOnPromptEnabled = boolValue
-		//case "CheckSensitiveOnCompletionEnabled":
-		//	constant.CheckSensitiveOnCompletionEnabled = boolValue
+		case "ModelRequestRateLimitEnabled":
+			setting.ModelRequestRateLimitEnabled = boolValue
 		case "StopOnSensitiveEnabled":
 			setting.StopOnSensitiveEnabled = boolValue
 		case "SMTPSSLEnabled":
@@ -304,6 +332,12 @@ func updateOptionMap(key string, value string) (err error) {
 		common.QuotaRemindThreshold, _ = strconv.Atoi(value)
 	case "PreConsumedQuota":
 		common.PreConsumedQuota, _ = strconv.Atoi(value)
+	case "ModelRequestRateLimitCount":
+		setting.ModelRequestRateLimitCount, _ = strconv.Atoi(value)
+	case "ModelRequestRateLimitDurationMinutes":
+		setting.ModelRequestRateLimitDurationMinutes, _ = strconv.Atoi(value)
+	case "ModelRequestRateLimitSuccessCount":
+		setting.ModelRequestRateLimitSuccessCount, _ = strconv.Atoi(value)
 	case "RetryTimes":
 		common.RetryTimes, _ = strconv.Atoi(value)
 	case "DataExportInterval":
@@ -311,29 +345,58 @@ func updateOptionMap(key string, value string) (err error) {
 	case "DataExportDefaultTime":
 		common.DataExportDefaultTime = value
 	case "ModelRatio":
-		err = common.UpdateModelRatioByJSONString(value)
+		err = operation_setting.UpdateModelRatioByJSONString(value)
 	case "GroupRatio":
 		err = setting.UpdateGroupRatioByJSONString(value)
 	case "UserUsableGroups":
 		err = setting.UpdateUserUsableGroupsByJSONString(value)
 	case "CompletionRatio":
-		err = common.UpdateCompletionRatioByJSONString(value)
+		err = operation_setting.UpdateCompletionRatioByJSONString(value)
 	case "ModelPrice":
-		err = common.UpdateModelPriceByJSONString(value)
+		err = operation_setting.UpdateModelPriceByJSONString(value)
+	case "CacheRatio":
+		err = operation_setting.UpdateCacheRatioByJSONString(value)
 	case "TopUpLink":
 		common.TopUpLink = value
-	case "ChatLink":
-		common.ChatLink = value
-	case "ChatLink2":
-		common.ChatLink2 = value
+	//case "ChatLink":
+	//	common.ChatLink = value
+	//case "ChatLink2":
+	//	common.ChatLink2 = value
 	case "ChannelDisableThreshold":
 		common.ChannelDisableThreshold, _ = strconv.ParseFloat(value, 64)
 	case "QuotaPerUnit":
 		common.QuotaPerUnit, _ = strconv.ParseFloat(value, 64)
 	case "SensitiveWords":
 		setting.SensitiveWordsFromString(value)
+	case "AutomaticDisableKeywords":
+		operation_setting.AutomaticDisableKeywordsFromString(value)
 	case "StreamCacheQueueLength":
 		setting.StreamCacheQueueLength, _ = strconv.Atoi(value)
 	}
 	return err
 }
+
+// handleConfigUpdate 处理分层配置更新，返回是否已处理
+func handleConfigUpdate(key, value string) bool {
+	parts := strings.SplitN(key, ".", 2)
+	if len(parts) != 2 {
+		return false // 不是分层配置
+	}
+
+	configName := parts[0]
+	configKey := parts[1]
+
+	// 获取配置对象
+	cfg := config.GlobalConfig.Get(configName)
+	if cfg == nil {
+		return false // 未注册的配置
+	}
+
+	// 更新配置
+	configMap := map[string]string{
+		configKey: value,
+	}
+	config.UpdateConfigFromMap(cfg, configMap)
+
+	return true // 已处理
+}

model/pricing.go

@@ -2,6 +2,7 @@ package model
 
 import (
 	"one-api/common"
+	"one-api/setting/operation_setting"
 	"sync"
 	"time"
 )
@@ -64,13 +65,14 @@ func updatePricing() {
 			ModelName:   model,
 			EnableGroup: groups,
 		}
-		modelPrice, findPrice := common.GetModelPrice(model, false)
+		modelPrice, findPrice := operation_setting.GetModelPrice(model, false)
 		if findPrice {
 			pricing.ModelPrice = modelPrice
 			pricing.QuotaType = 1
 		} else {
-			pricing.ModelRatio = common.GetModelRatio(model)
-			pricing.CompletionRatio = common.GetCompletionRatio(model)
+			modelRatio, _ := operation_setting.GetModelRatio(model)
+			pricing.ModelRatio = modelRatio
+			pricing.CompletionRatio = operation_setting.GetCompletionRatio(model)
 			pricing.QuotaType = 0
 		}
 		pricingMap = append(pricingMap, pricing)

model/token.go

@@ -3,13 +3,11 @@ package model
 import (
 	"errors"
 	"fmt"
+	"one-api/common"
+	"strings"
+
 	"github.com/bytedance/gopkg/util/gopool"
 	"gorm.io/gorm"
-	"one-api/common"
-	relaycommon "one-api/relay/common"
-	"one-api/setting"
-	"strconv"
-	"strings"
 )
 
 type Token struct {
@@ -322,80 +320,3 @@ func decreaseTokenQuota(id int, quota int) (err error) {
 	).Error
 	return err
 }
-
-func PreConsumeTokenQuota(relayInfo *relaycommon.RelayInfo, quota int) error {
-	if quota < 0 {
-		return errors.New("quota 不能为负数！")
-	}
-	if relayInfo.IsPlayground {
-		return nil
-	}
-	//if relayInfo.TokenUnlimited {
-	//	return nil
-	//}
-	token, err := GetTokenById(relayInfo.TokenId)
-	if err != nil {
-		return err
-	}
-	if !relayInfo.TokenUnlimited && token.RemainQuota < quota {
-		return errors.New("令牌额度不足")
-	}
-	err = DecreaseTokenQuota(relayInfo.TokenId, relayInfo.TokenKey, quota)
-	if err != nil {
-		return err
-	}
-	return nil
-}
-
-func PostConsumeQuota(relayInfo *relaycommon.RelayInfo, userQuota int, quota int, preConsumedQuota int, sendEmail bool) (err error) {
-
-	if quota > 0 {
-		err = DecreaseUserQuota(relayInfo.UserId, quota)
-	} else {
-		err = IncreaseUserQuota(relayInfo.UserId, -quota)
-	}
-	if err != nil {
-		return err
-	}
-
-	if !relayInfo.IsPlayground {
-		if quota > 0 {
-			err = DecreaseTokenQuota(relayInfo.TokenId, relayInfo.TokenKey, quota)
-		} else {
-			err = IncreaseTokenQuota(relayInfo.TokenId, relayInfo.TokenKey, -quota)
-		}
-		if err != nil {
-			return err
-		}
-	}
-
-	if sendEmail {
-		if (quota + preConsumedQuota) != 0 {
-			quotaTooLow := userQuota >= common.QuotaRemindThreshold && userQuota-(quota+preConsumedQuota) < common.QuotaRemindThreshold
-			noMoreQuota := userQuota-(quota+preConsumedQuota) <= 0
-			if quotaTooLow || noMoreQuota {
-				go func() {
-					email, err := GetUserEmail(relayInfo.UserId)
-					if err != nil {
-						common.SysError("failed to fetch user email: " + err.Error())
-					}
-					prompt := "您的额度即将用尽"
-					if noMoreQuota {
-						prompt = "您的额度已用尽"
-					}
-					if email != "" {
-						topUpLink := fmt.Sprintf("%s/topup", setting.ServerAddress)
-						err = common.SendEmail(prompt, email,
-							fmt.Sprintf("%s，当前剩余额度为 %d，为了不影响您的使用，请及时充值。<br/>充值链接：<a href='%s'>%s</a>", prompt, userQuota, topUpLink, topUpLink))
-						if err != nil {
-							common.SysError("failed to send email" + err.Error())
-						}
-						common.SysLog("user quota is low, consumed quota: " + strconv.Itoa(quota) + ", user quota: " + strconv.Itoa(userQuota))
-					}
-				}()
-			}
-		}
-	}
-
-	return nil
-}

model/token_cache.go

@@ -52,7 +52,7 @@ func cacheSetTokenField(key string, field string, value string) error {
 func cacheGetTokenByKey(key string) (*Token, error) {
 	hmacKey := common.GenerateHMAC(key)
 	if !common.RedisEnabled {
-		return nil, nil
+		return nil, fmt.Errorf("redis is not enabled")
 	}
 	var token Token
 	err := common.RedisHGetObj(fmt.Sprintf("token:%s", hmacKey), &token)

model/topup.go

@@ -3,7 +3,7 @@ package model
 type TopUp struct {
 	Id         int     `json:"id"`
 	UserId     int     `json:"user_id" gorm:"index"`
-	Amount     int     `json:"amount"`
+	Amount     int64   `json:"amount"`
 	Money      float64 `json:"money"`
 	TradeNo    string  `json:"trade_no"`
 	CreateTime int64   `json:"create_time"`

model/user.go

@@ -1,6 +1,7 @@
 package model
 
 import (
+	"encoding/json"
 	"errors"
 	"fmt"
 	"one-api/common"
@@ -38,6 +39,20 @@ type User struct {
 	InviterId        int            `json:"inviter_id" gorm:"type:int;column:inviter_id;index"`
 	DeletedAt        gorm.DeletedAt `gorm:"index"`
 	LinuxDOId        string         `json:"linux_do_id" gorm:"column:linux_do_id;index"`
+	Setting          string         `json:"setting" gorm:"type:text;column:setting"`
+}
+
+func (user *User) ToBaseUser() *UserBase {
+	cache := &UserBase{
+		Id:       user.Id,
+		Group:    user.Group,
+		Quota:    user.Quota,
+		Status:   user.Status,
+		Username: user.Username,
+		Setting:  user.Setting,
+		Email:    user.Email,
+	}
+	return cache
 }
 
 func (user *User) GetAccessToken() string {
@@ -51,6 +66,22 @@ func (user *User) SetAccessToken(token string) {
 	user.AccessToken = &token
 }
 
+func (user *User) GetSetting() map[string]interface{} {
+	if user.Setting == "" {
+		return nil
+	}
+	return common.StrToMap(user.Setting)
+}
+
+func (user *User) SetSetting(setting map[string]interface{}) {
+	settingBytes, err := json.Marshal(setting)
+	if err != nil {
+		common.SysError("failed to marshal setting: " + err.Error())
+		return
+	}
+	user.Setting = string(settingBytes)
+}
+
 // CheckUserExistOrDeleted check if user exist or deleted, if not exist, return false, nil, if deleted or exist, return true, nil
 func CheckUserExistOrDeleted(username string, email string) (bool, error) {
 	var user User
@@ -289,7 +320,7 @@ func (user *User) Insert(inviterId int) error {
 	}
 	if inviterId != 0 {
 		if common.QuotaForInvitee > 0 {
-			_ = IncreaseUserQuota(user.Id, common.QuotaForInvitee)
+			_ = IncreaseUserQuota(user.Id, common.QuotaForInvitee, true)
 			RecordLog(user.Id, LogTypeSystem, fmt.Sprintf("使用邀请码赠送 %s", common.LogQuota(common.QuotaForInvitee)))
 		}
 		if common.QuotaForInviter > 0 {
@@ -315,8 +346,8 @@ func (user *User) Update(updatePassword bool) error {
 		return err
 	}
 
-	// 更新缓存
-	return updateUserCache(user.Id, user.Username, user.Group, user.Quota, user.Status)
+	// Update cache
+	return updateUserCache(*user)
 }
 
 func (user *User) Edit(updatePassword bool) error {
@@ -344,8 +375,8 @@ func (user *User) Edit(updatePassword bool) error {
 		return err
 	}
 
-	// 更新缓存
-	return updateUserCache(user.Id, user.Username, user.Group, user.Quota, user.Status)
+	// Update cache
+	return updateUserCache(*user)
 }
 
 func (user *User) Delete() error {
@@ -371,8 +402,8 @@ func (user *User) HardDelete() error {
 // ValidateAndFill check password & user status
 func (user *User) ValidateAndFill() (err error) {
 	// When querying with struct, GORM will only query with non-zero fields,
-	// that means if your field’s value is 0, '', false or other zero values,
-	// it won’t be used to build query conditions
+	// that means if your field's value is 0, '', false or other zero values,
+	// it won't be used to build query conditions
 	password := user.Password
 	username := strings.TrimSpace(user.Username)
 	if username == "" || password == "" {
@@ -471,35 +502,35 @@ func IsAdmin(userId int) bool {
 	return user.Role >= common.RoleAdminUser
 }
 
-// IsUserEnabled checks user status from Redis first, falls back to DB if needed
-func IsUserEnabled(id int, fromDB bool) (status bool, err error) {
-	defer func() {
-		// Update Redis cache asynchronously on successful DB read
-		if shouldUpdateRedis(fromDB, err) {
-			gopool.Go(func() {
-				if err := updateUserStatusCache(id, status); err != nil {
-					common.SysError("failed to update user status cache: " + err.Error())
-				}
-			})
-		}
-	}()
-	if !fromDB && common.RedisEnabled {
-		// Try Redis first
-		status, err := getUserStatusCache(id)
-		if err == nil {
-			return status == common.UserStatusEnabled, nil
-		}
-		// Don't return error - fall through to DB
-	}
-	fromDB = true
-	var user User
-	err = DB.Where("id = ?", id).Select("status").Find(&user).Error
-	if err != nil {
-		return false, err
-	}
-
-	return user.Status == common.UserStatusEnabled, nil
-}
+//// IsUserEnabled checks user status from Redis first, falls back to DB if needed
+//func IsUserEnabled(id int, fromDB bool) (status bool, err error) {
+//	defer func() {
+//		// Update Redis cache asynchronously on successful DB read
+//		if shouldUpdateRedis(fromDB, err) {
+//			gopool.Go(func() {
+//				if err := updateUserStatusCache(id, status); err != nil {
+//					common.SysError("failed to update user status cache: " + err.Error())
+//				}
+//			})
+//		}
+//	}()
+//	if !fromDB && common.RedisEnabled {
+//		// Try Redis first
+//		status, err := getUserStatusCache(id)
+//		if err == nil {
+//			return status == common.UserStatusEnabled, nil
+//		}
+//		// Don't return error - fall through to DB
+//	}
+//	fromDB = true
+//	var user User
+//	err = DB.Where("id = ?", id).Select("status").Find(&user).Error
+//	if err != nil {
+//		return false, err
+//	}
+//
+//	return user.Status == common.UserStatusEnabled, nil
+//}
 
 func ValidateAccessToken(token string) (user *User) {
 	if token == "" {
@@ -531,7 +562,6 @@ func GetUserQuota(id int, fromDB bool) (quota int, err error) {
 			return quota, nil
 		}
 		// Don't return error - fall through to DB
-		//common.SysError("failed to get user quota from cache: " + err.Error())
 	}
 	fromDB = true
 	err = DB.Model(&User{}).Where("id = ?", id).Select("quota").Find(&quota).Error
@@ -580,7 +610,36 @@ func GetUserGroup(id int, fromDB bool) (group string, err error) {
 	return group, nil
 }
 
-func IncreaseUserQuota(id int, quota int) (err error) {
+// GetUserSetting gets setting from Redis first, falls back to DB if needed
+func GetUserSetting(id int, fromDB bool) (settingMap map[string]interface{}, err error) {
+	var setting string
+	defer func() {
+		// Update Redis cache asynchronously on successful DB read
+		if shouldUpdateRedis(fromDB, err) {
+			gopool.Go(func() {
+				if err := updateUserSettingCache(id, setting); err != nil {
+					common.SysError("failed to update user setting cache: " + err.Error())
+				}
+			})
+		}
+	}()
+	if !fromDB && common.RedisEnabled {
+		setting, err := getUserSettingCache(id)
+		if err == nil {
+			return setting, nil
+		}
+		// Don't return error - fall through to DB
+	}
+	fromDB = true
+	err = DB.Model(&User{}).Where("id = ?", id).Select("setting").Find(&setting).Error
+	if err != nil {
+		return map[string]interface{}{}, err
+	}
+
+	return common.StrToMap(setting), nil
+}
+
+func IncreaseUserQuota(id int, quota int, db bool) (err error) {
 	if quota < 0 {
 		return errors.New("quota 不能为负数！")
 	}
@@ -590,7 +649,7 @@ func IncreaseUserQuota(id int, quota int) (err error) {
 			common.SysError("failed to increase user quota: " + err.Error())
 		}
 	})
-	if common.BatchUpdateEnabled {
+	if !db && common.BatchUpdateEnabled {
 		addNewRecord(BatchUpdateTypeUserQuota, id, quota)
 		return nil
 	}
@@ -635,15 +694,20 @@ func DeltaUpdateUserQuota(id int, delta int) (err error) {
 		return nil
 	}
 	if delta > 0 {
-		return IncreaseUserQuota(id, delta)
+		return IncreaseUserQuota(id, delta, false)
 	} else {
 		return DecreaseUserQuota(id, -delta)
 	}
 }
 
-func GetRootUserEmail() (email string) {
-	DB.Model(&User{}).Where("role = ?", common.RoleRootUser).Select("email").Find(&email)
-	return email
+//func GetRootUserEmail() (email string) {
+//	DB.Model(&User{}).Where("role = ?", common.RoleRootUser).Select("email").Find(&email)
+//	return email
+//}
+
+func GetRootUser() (user *User) {
+	DB.Where("role = ?", common.RoleRootUser).First(&user)
+	return user
 }
 
 func UpdateUserUsedQuotaAndRequestCount(id int, quota int) {
@@ -725,10 +789,10 @@ func IsLinuxDOIdAlreadyTaken(linuxDOId string) bool {
 	return !errors.Is(err, gorm.ErrRecordNotFound)
 }
 
-func (u *User) FillUserByLinuxDOId() error {
-	if u.LinuxDOId == "" {
+func (user *User) FillUserByLinuxDOId() error {
+	if user.LinuxDOId == "" {
 		return errors.New("linux do id is empty")
 	}
-	err := DB.Where("linux_do_id = ?", u.LinuxDOId).First(u).Error
+	err := DB.Where("linux_do_id = ?", user.LinuxDOId).First(user).Error
 	return err
 }

model/user_cache.go

@@ -1,206 +1,223 @@
 package model
 
 import (
+	"encoding/json"
 	"fmt"
+	"github.com/gin-gonic/gin"
 	"one-api/common"
 	"one-api/constant"
-	"strconv"
 	"time"
+
+	"github.com/bytedance/gopkg/util/gopool"
 )
 
-// Change UserCache struct to userCache
-type userCache struct {
+// UserBase struct remains the same as it represents the cached data structure
+type UserBase struct {
 	Id       int    `json:"id"`
 	Group    string `json:"group"`
+	Email    string `json:"email"`
 	Quota    int    `json:"quota"`
 	Status   int    `json:"status"`
-	Role     int    `json:"role"`
 	Username string `json:"username"`
+	Setting  string `json:"setting"`
 }
 
-// Rename all exported functions to private ones
-// invalidateUserCache clears all user related cache
+func (user *UserBase) WriteContext(c *gin.Context) {
+	c.Set(constant.ContextKeyUserGroup, user.Group)
+	c.Set(constant.ContextKeyUserQuota, user.Quota)
+	c.Set(constant.ContextKeyUserStatus, user.Status)
+	c.Set(constant.ContextKeyUserEmail, user.Email)
+	c.Set("username", user.Username)
+	c.Set(constant.ContextKeyUserSetting, user.GetSetting())
+}
+
+func (user *UserBase) GetSetting() map[string]interface{} {
+	if user.Setting == "" {
+		return nil
+	}
+	return common.StrToMap(user.Setting)
+}
+
+func (user *UserBase) SetSetting(setting map[string]interface{}) {
+	settingBytes, err := json.Marshal(setting)
+	if err != nil {
+		common.SysError("failed to marshal setting: " + err.Error())
+		return
+	}
+	user.Setting = string(settingBytes)
+}
+
+// getUserCacheKey returns the key for user cache
+func getUserCacheKey(userId int) string {
+	return fmt.Sprintf("user:%d", userId)
+}
+
+// invalidateUserCache clears user cache
 func invalidateUserCache(userId int) error {
 	if !common.RedisEnabled {
 		return nil
 	}
+	return common.RedisHDelObj(getUserCacheKey(userId))
+}
 
-	keys := []string{
-		fmt.Sprintf(constant.UserGroupKeyFmt, userId),
-		fmt.Sprintf(constant.UserQuotaKeyFmt, userId),
-		fmt.Sprintf(constant.UserEnabledKeyFmt, userId),
-		fmt.Sprintf(constant.UserUsernameKeyFmt, userId),
+// updateUserCache updates all user cache fields using hash
+func updateUserCache(user User) error {
+	if !common.RedisEnabled {
+		return nil
 	}
 
-	for _, key := range keys {
-		if err := common.RedisDel(key); err != nil {
-			return fmt.Errorf("failed to delete cache key %s: %w", key, err)
+	return common.RedisHSetObj(
+		getUserCacheKey(user.Id),
+		user.ToBaseUser(),
+		time.Duration(constant.UserId2QuotaCacheSeconds)*time.Second,
+	)
+}
+
+// GetUserCache gets complete user cache from hash
+func GetUserCache(userId int) (userCache *UserBase, err error) {
+	var user *User
+	var fromDB bool
+	defer func() {
+		// Update Redis cache asynchronously on successful DB read
+		if shouldUpdateRedis(fromDB, err) && user != nil {
+			gopool.Go(func() {
+				if err := updateUserCache(*user); err != nil {
+					common.SysError("failed to update user status cache: " + err.Error())
+				}
+			})
 		}
-	}
-	return nil
-}
+	}()
 
-// updateUserGroupCache updates user group cache
-func updateUserGroupCache(userId int, group string) error {
-	if !common.RedisEnabled {
-		return nil
-	}
-	return common.RedisSet(
-		fmt.Sprintf(constant.UserGroupKeyFmt, userId),
-		group,
-		time.Duration(constant.UserId2QuotaCacheSeconds)*time.Second,
-	)
-}
-
-// updateUserQuotaCache updates user quota cache
-func updateUserQuotaCache(userId int, quota int) error {
-	if !common.RedisEnabled {
-		return nil
-	}
-	return common.RedisSet(
-		fmt.Sprintf(constant.UserQuotaKeyFmt, userId),
-		fmt.Sprintf("%d", quota),
-		time.Duration(constant.UserId2QuotaCacheSeconds)*time.Second,
-	)
-}
-
-// updateUserStatusCache updates user status cache
-func updateUserStatusCache(userId int, userEnabled bool) error {
-	if !common.RedisEnabled {
-		return nil
-	}
-	enabled := "0"
-	if userEnabled {
-		enabled = "1"
-	}
-	return common.RedisSet(
-		fmt.Sprintf(constant.UserEnabledKeyFmt, userId),
-		enabled,
-		time.Duration(constant.UserId2StatusCacheSeconds)*time.Second,
-	)
-}
-
-// updateUserNameCache updates username cache
-func updateUserNameCache(userId int, username string) error {
-	if !common.RedisEnabled {
-		return nil
-	}
-	return common.RedisSet(
-		fmt.Sprintf(constant.UserUsernameKeyFmt, userId),
-		username,
-		time.Duration(constant.UserId2QuotaCacheSeconds)*time.Second,
-	)
-}
-
-// updateUserCache updates all user cache fields
-func updateUserCache(userId int, username string, userGroup string, quota int, status int) error {
-	if !common.RedisEnabled {
-		return nil
+	// Try getting from Redis first
+	userCache, err = cacheGetUserBase(userId)
+	if err == nil {
+		return userCache, nil
 	}
 
-	if err := updateUserGroupCache(userId, userGroup); err != nil {
-		return fmt.Errorf("update group cache: %w", err)
-	}
-
-	if err := updateUserQuotaCache(userId, quota); err != nil {
-		return fmt.Errorf("update quota cache: %w", err)
-	}
-
-	if err := updateUserStatusCache(userId, status == common.UserStatusEnabled); err != nil {
-		return fmt.Errorf("update status cache: %w", err)
-	}
-
-	if err := updateUserNameCache(userId, username); err != nil {
-		return fmt.Errorf("update username cache: %w", err)
-	}
-
-	return nil
-}
-
-// getUserGroupCache gets user group from cache
-func getUserGroupCache(userId int) (string, error) {
-	if !common.RedisEnabled {
-		return "", nil
-	}
-	return common.RedisGet(fmt.Sprintf(constant.UserGroupKeyFmt, userId))
-}
-
-// getUserQuotaCache gets user quota from cache
-func getUserQuotaCache(userId int) (int, error) {
-	if !common.RedisEnabled {
-		return 0, nil
-	}
-	quotaStr, err := common.RedisGet(fmt.Sprintf(constant.UserQuotaKeyFmt, userId))
+	// If Redis fails, get from DB
+	fromDB = true
+	user, err = GetUserById(userId, false)
 	if err != nil {
-		return 0, err
+		return nil, err // Return nil and error if DB lookup fails
 	}
-	return strconv.Atoi(quotaStr)
+
+	// Create cache object from user data
+	userCache = &UserBase{
+		Id:       user.Id,
+		Group:    user.Group,
+		Quota:    user.Quota,
+		Status:   user.Status,
+		Username: user.Username,
+		Setting:  user.Setting,
+		Email:    user.Email,
+	}
+
+	return userCache, nil
 }
 
-// getUserStatusCache gets user status from cache
-func getUserStatusCache(userId int) (int, error) {
+func cacheGetUserBase(userId int) (*UserBase, error) {
 	if !common.RedisEnabled {
-		return 0, nil
+		return nil, fmt.Errorf("redis is not enabled")
 	}
-	statusStr, err := common.RedisGet(fmt.Sprintf(constant.UserEnabledKeyFmt, userId))
+	var userCache UserBase
+	// Try getting from Redis first
+	err := common.RedisHGetObj(getUserCacheKey(userId), &userCache)
 	if err != nil {
-		return 0, err
+		return nil, err
 	}
-	return strconv.Atoi(statusStr)
+	return &userCache, nil
 }
 
-// getUserNameCache gets username from cache
-func getUserNameCache(userId int) (string, error) {
-	if !common.RedisEnabled {
-		return "", nil
-	}
-	return common.RedisGet(fmt.Sprintf(constant.UserUsernameKeyFmt, userId))
-}
-
-// getUserCache gets complete user cache
-func getUserCache(userId int) (*userCache, error) {
-	if !common.RedisEnabled {
-		return nil, nil
-	}
-
-	group, err := getUserGroupCache(userId)
-	if err != nil {
-		return nil, fmt.Errorf("get group cache: %w", err)
-	}
-
-	quota, err := getUserQuotaCache(userId)
-	if err != nil {
-		return nil, fmt.Errorf("get quota cache: %w", err)
-	}
-
-	status, err := getUserStatusCache(userId)
-	if err != nil {
-		return nil, fmt.Errorf("get status cache: %w", err)
-	}
-
-	username, err := getUserNameCache(userId)
-	if err != nil {
-		return nil, fmt.Errorf("get username cache: %w", err)
-	}
-
-	return &userCache{
-		Id:       userId,
-		Group:    group,
-		Quota:    quota,
-		Status:   status,
-		Username: username,
-	}, nil
-}
-
-// Add atomic quota operations
+// Add atomic quota operations using hash fields
 func cacheIncrUserQuota(userId int, delta int64) error {
 	if !common.RedisEnabled {
 		return nil
 	}
-	key := fmt.Sprintf(constant.UserQuotaKeyFmt, userId)
-	return common.RedisIncr(key, delta)
+	return common.RedisHIncrBy(getUserCacheKey(userId), "Quota", delta)
 }
 
 func cacheDecrUserQuota(userId int, delta int64) error {
 	return cacheIncrUserQuota(userId, -delta)
 }
+
+// Helper functions to get individual fields if needed
+func getUserGroupCache(userId int) (string, error) {
+	cache, err := GetUserCache(userId)
+	if err != nil {
+		return "", err
+	}
+	return cache.Group, nil
+}
+
+func getUserQuotaCache(userId int) (int, error) {
+	cache, err := GetUserCache(userId)
+	if err != nil {
+		return 0, err
+	}
+	return cache.Quota, nil
+}
+
+func getUserStatusCache(userId int) (int, error) {
+	cache, err := GetUserCache(userId)
+	if err != nil {
+		return 0, err
+	}
+	return cache.Status, nil
+}
+
+func getUserNameCache(userId int) (string, error) {
+	cache, err := GetUserCache(userId)
+	if err != nil {
+		return "", err
+	}
+	return cache.Username, nil
+}
+
+func getUserSettingCache(userId int) (map[string]interface{}, error) {
+	setting := make(map[string]interface{})
+	cache, err := GetUserCache(userId)
+	if err != nil {
+		return setting, err
+	}
+	return cache.GetSetting(), nil
+}
+
+// New functions for individual field updates
+func updateUserStatusCache(userId int, status bool) error {
+	if !common.RedisEnabled {
+		return nil
+	}
+	statusInt := common.UserStatusEnabled
+	if !status {
+		statusInt = common.UserStatusDisabled
+	}
+	return common.RedisHSetField(getUserCacheKey(userId), "Status", fmt.Sprintf("%d", statusInt))
+}
+
+func updateUserQuotaCache(userId int, quota int) error {
+	if !common.RedisEnabled {
+		return nil
+	}
+	return common.RedisHSetField(getUserCacheKey(userId), "Quota", fmt.Sprintf("%d", quota))
+}
+
+func updateUserGroupCache(userId int, group string) error {
+	if !common.RedisEnabled {
+		return nil
+	}
+	return common.RedisHSetField(getUserCacheKey(userId), "Group", group)
+}
+
+func updateUserNameCache(userId int, username string) error {
+	if !common.RedisEnabled {
+		return nil
+	}
+	return common.RedisHSetField(getUserCacheKey(userId), "Username", username)
+}
+
+func updateUserSettingCache(userId int, setting string) error {
+	if !common.RedisEnabled {
+		return nil
+	}
+	return common.RedisHSetField(getUserCacheKey(userId), "Setting", setting)
+}

relay/channel/adapter.go

@@ -15,6 +15,7 @@ type Adaptor interface {
 	SetupRequestHeader(c *gin.Context, req *http.Header, info *relaycommon.RelayInfo) error
 	ConvertRequest(c *gin.Context, info *relaycommon.RelayInfo, request *dto.GeneralOpenAIRequest) (any, error)
 	ConvertRerankRequest(c *gin.Context, relayMode int, request dto.RerankRequest) (any, error)
+	ConvertEmbeddingRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.EmbeddingRequest) (any, error)
 	ConvertAudioRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.AudioRequest) (io.Reader, error)
 	ConvertImageRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.ImageRequest) (any, error)
 	DoRequest(c *gin.Context, info *relaycommon.RelayInfo, requestBody io.Reader) (any, error)

relay/channel/ali/adaptor.go

@@ -49,9 +49,6 @@ func (a *Adaptor) ConvertRequest(c *gin.Context, info *relaycommon.RelayInfo, re
 		return nil, errors.New("request is nil")
 	}
 	switch info.RelayMode {
-	case constant.RelayModeEmbeddings:
-		baiduEmbeddingRequest := embeddingRequestOpenAI2Ali(*request)
-		return baiduEmbeddingRequest, nil
 	default:
 		aliReq := requestOpenAI2Ali(*request)
 		return aliReq, nil
@@ -67,6 +64,10 @@ func (a *Adaptor) ConvertRerankRequest(c *gin.Context, relayMode int, request dt
 	return nil, errors.New("not implemented")
 }
 
+func (a *Adaptor) ConvertEmbeddingRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.EmbeddingRequest) (any, error) {
+	return embeddingRequestOpenAI2Ali(request), nil
+}
+
 func (a *Adaptor) ConvertAudioRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.AudioRequest) (io.Reader, error) {
 	//TODO implement me
 	return nil, errors.New("not implemented")

relay/channel/ali/image.go

@@ -27,7 +27,7 @@ func oaiImage2Ali(request dto.ImageRequest) *AliImageRequest {
 }
 
 func updateTask(info *relaycommon.RelayInfo, taskID string, key string) (*AliResponse, error, []byte) {
-	url := fmt.Sprintf("/api/v1/tasks/%s", taskID)
+	url := fmt.Sprintf("%s/api/v1/tasks/%s", info.BaseUrl, taskID)
 
 	var aliResponse AliResponse
 

relay/channel/ali/text.go

@@ -8,6 +8,7 @@ import (
 	"net/http"
 	"one-api/common"
 	"one-api/dto"
+	"one-api/relay/helper"
 	"one-api/service"
 	"strings"
 )
@@ -25,9 +26,12 @@ func requestOpenAI2Ali(request dto.GeneralOpenAIRequest) *dto.GeneralOpenAIReque
 	return &request
 }
 
-func embeddingRequestOpenAI2Ali(request dto.GeneralOpenAIRequest) *AliEmbeddingRequest {
+func embeddingRequestOpenAI2Ali(request dto.EmbeddingRequest) *AliEmbeddingRequest {
+	if request.Model == "" {
+		request.Model = "text-embedding-v1"
+	}
 	return &AliEmbeddingRequest{
-		Model: "text-embedding-v1",
+		Model: request.Model,
 		Input: struct {
 			Texts []string `json:"texts"`
 		}{
@@ -150,7 +154,7 @@ func aliStreamHandler(c *gin.Context, resp *http.Response) (*dto.OpenAIErrorWith
 		}
 		stopChan <- true
 	}()
-	service.SetEventStreamHeaders(c)
+	helper.SetEventStreamHeaders(c)
 	lastResponseText := ""
 	c.Stream(func(w io.Writer) bool {
 		select {

relay/channel/api_request.go

@@ -39,7 +39,7 @@ func DoApiRequest(a Adaptor, c *gin.Context, info *common.RelayInfo, requestBody
 	if err != nil {
 		return nil, fmt.Errorf("setup request header failed: %w", err)
 	}
-	resp, err := doRequest(c, req)
+	resp, err := doRequest(c, req, info)
 	if err != nil {
 		return nil, fmt.Errorf("do request failed: %w", err)
 	}
@@ -62,7 +62,7 @@ func DoFormRequest(a Adaptor, c *gin.Context, info *common.RelayInfo, requestBod
 	if err != nil {
 		return nil, fmt.Errorf("setup request header failed: %w", err)
 	}
-	resp, err := doRequest(c, req)
+	resp, err := doRequest(c, req, info)
 	if err != nil {
 		return nil, fmt.Errorf("do request failed: %w", err)
 	}
@@ -90,8 +90,18 @@ func DoWssRequest(a Adaptor, c *gin.Context, info *common.RelayInfo, requestBody
 	return targetConn, nil
 }
 
-func doRequest(c *gin.Context, req *http.Request) (*http.Response, error) {
-	resp, err := service.GetHttpClient().Do(req)
+func doRequest(c *gin.Context, req *http.Request, info *common.RelayInfo) (*http.Response, error) {
+	var client *http.Client
+	var err error
+	if proxyURL, ok := info.ChannelSetting["proxy"]; ok {
+		client, err = service.NewProxyHttpClient(proxyURL.(string))
+		if err != nil {
+			return nil, fmt.Errorf("new proxy http client failed: %w", err)
+		}
+	} else {
+		client = service.GetHttpClient()
+	}
+	resp, err := client.Do(req)
 	if err != nil {
 		return nil, err
 	}
@@ -120,7 +130,7 @@ func DoTaskApiRequest(a TaskAdaptor, c *gin.Context, info *common.TaskRelayInfo,
 	if err != nil {
 		return nil, fmt.Errorf("setup request header failed: %w", err)
 	}
-	resp, err := doRequest(c, req)
+	resp, err := doRequest(c, req, info.RelayInfo)
 	if err != nil {
 		return nil, fmt.Errorf("do request failed: %w", err)
 	}

relay/channel/aws/adaptor.go

@@ -8,6 +8,7 @@ import (
 	"one-api/dto"
 	"one-api/relay/channel/claude"
 	relaycommon "one-api/relay/common"
+	"one-api/setting/model_setting"
 )
 
 const (
@@ -38,6 +39,7 @@ func (a *Adaptor) GetRequestURL(info *relaycommon.RelayInfo) (string, error) {
 }
 
 func (a *Adaptor) SetupRequestHeader(c *gin.Context, req *http.Header, info *relaycommon.RelayInfo) error {
+	model_setting.GetClaudeSettings().WriteHeaders(info.OriginModelName, req)
 	return nil
 }
 
@@ -49,8 +51,10 @@ func (a *Adaptor) ConvertRequest(c *gin.Context, info *relaycommon.RelayInfo, re
 	var claudeReq *claude.ClaudeRequest
 	var err error
 	claudeReq, err = claude.RequestOpenAI2ClaudeMessage(*request)
-
-	c.Set("request_model", request.Model)
+	if err != nil {
+		return nil, err
+	}
+	c.Set("request_model", claudeReq.Model)
 	c.Set("converted_request", claudeReq)
 	return claudeReq, err
 }
@@ -59,6 +63,11 @@ func (a *Adaptor) ConvertRerankRequest(c *gin.Context, relayMode int, request dt
 	return nil, nil
 }
 
+func (a *Adaptor) ConvertEmbeddingRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.EmbeddingRequest) (any, error) {
+	//TODO implement me
+	return nil, errors.New("not implemented")
+}
+
 func (a *Adaptor) DoRequest(c *gin.Context, info *relaycommon.RelayInfo, requestBody io.Reader) (any, error) {
 	return nil, nil
 }

relay/channel/aws/constants.go

@@ -9,7 +9,8 @@ var awsModelIDMap = map[string]string{
 	"claude-3-haiku-20240307":    "anthropic.claude-3-haiku-20240307-v1:0",
 	"claude-3-5-sonnet-20240620": "anthropic.claude-3-5-sonnet-20240620-v1:0",
 	"claude-3-5-sonnet-20241022": "anthropic.claude-3-5-sonnet-20241022-v2:0",
-	"claude-3-5-haiku-20241022": "anthropic.claude-3-5-haiku-20241022-v1:0",
+	"claude-3-5-haiku-20241022":  "anthropic.claude-3-5-haiku-20241022-v1:0",
+	"claude-3-7-sonnet-20250219": "anthropic.claude-3-7-sonnet-20250219-v1:0",
 }
 
 var ChannelName = "aws"

relay/channel/aws/dto.go

@@ -14,8 +14,9 @@ type AwsClaudeRequest struct {
 	TopP             float64                `json:"top_p,omitempty"`
 	TopK             int                    `json:"top_k,omitempty"`
 	StopSequences    []string               `json:"stop_sequences,omitempty"`
-	Tools            []claude.Tool          `json:"tools,omitempty"`
+	Tools            any                    `json:"tools,omitempty"`
 	ToolChoice       any                    `json:"tool_choice,omitempty"`
+	Thinking         *claude.Thinking       `json:"thinking,omitempty"`
 }
 
 func copyRequest(req *claude.ClaudeRequest) *AwsClaudeRequest {
@@ -30,5 +31,6 @@ func copyRequest(req *claude.ClaudeRequest) *AwsClaudeRequest {
 		StopSequences:    req.StopSequences,
 		Tools:            req.Tools,
 		ToolChoice:       req.ToolChoice,
+		Thinking:         req.Thinking,
 	}
 }

relay/channel/aws/relay-aws.go

@@ -12,6 +12,7 @@ import (
 	relaymodel "one-api/dto"
 	"one-api/relay/channel/claude"
 	relaycommon "one-api/relay/common"
+	"one-api/relay/helper"
 	"one-api/service"
 	"strings"
 	"time"
@@ -203,13 +204,13 @@ func awsStreamHandler(c *gin.Context, resp *http.Response, info *relaycommon.Rel
 		}
 	})
 	if info.ShouldIncludeUsage {
-		response := service.GenerateFinalUsageResponse(id, createdTime, info.UpstreamModelName, usage)
-		err := service.ObjectData(c, response)
+		response := helper.GenerateFinalUsageResponse(id, createdTime, info.UpstreamModelName, usage)
+		err := helper.ObjectData(c, response)
 		if err != nil {
 			common.SysError("send final response failed: " + err.Error())
 		}
 	}
-	service.Done(c)
+	helper.Done(c)
 	if resp != nil {
 		err = resp.Body.Close()
 		if err != nil {

relay/channel/baidu/adaptor.go

@@ -109,9 +109,6 @@ func (a *Adaptor) ConvertRequest(c *gin.Context, info *relaycommon.RelayInfo, re
 		return nil, errors.New("request is nil")
 	}
 	switch info.RelayMode {
-	case constant.RelayModeEmbeddings:
-		baiduEmbeddingRequest := embeddingRequestOpenAI2Baidu(*request)
-		return baiduEmbeddingRequest, nil
 	default:
 		baiduRequest := requestOpenAI2Baidu(*request)
 		return baiduRequest, nil
@@ -122,6 +119,11 @@ func (a *Adaptor) ConvertRerankRequest(c *gin.Context, relayMode int, request dt
 	return nil, nil
 }
 
+func (a *Adaptor) ConvertEmbeddingRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.EmbeddingRequest) (any, error) {
+	baiduEmbeddingRequest := embeddingRequestOpenAI2Baidu(request)
+	return baiduEmbeddingRequest, nil
+}
+
 func (a *Adaptor) DoRequest(c *gin.Context, info *relaycommon.RelayInfo, requestBody io.Reader) (any, error) {
 	return channel.DoApiRequest(a, c, info, requestBody)
 }

relay/channel/baidu/relay-baidu.go

@@ -11,6 +11,7 @@ import (
 	"one-api/common"
 	"one-api/constant"
 	"one-api/dto"
+	"one-api/relay/helper"
 	"one-api/service"
 	"strings"
 	"sync"
@@ -87,7 +88,7 @@ func streamResponseBaidu2OpenAI(baiduResponse *BaiduChatStreamResponse) *dto.Cha
 	return &response
 }
 
-func embeddingRequestOpenAI2Baidu(request dto.GeneralOpenAIRequest) *BaiduEmbeddingRequest {
+func embeddingRequestOpenAI2Baidu(request dto.EmbeddingRequest) *BaiduEmbeddingRequest {
 	return &BaiduEmbeddingRequest{
 		Input: request.ParseInput(),
 	}
@@ -138,7 +139,7 @@ func baiduStreamHandler(c *gin.Context, resp *http.Response) (*dto.OpenAIErrorWi
 		}
 		stopChan <- true
 	}()
-	service.SetEventStreamHeaders(c)
+	helper.SetEventStreamHeaders(c)
 	c.Stream(func(w io.Writer) bool {
 		select {
 		case data := <-dataChan:

relay/channel/baidu_v2/adaptor.go

@@ -0,0 +1,76 @@
+package baidu_v2
+
+import (
+	"errors"
+	"fmt"
+	"github.com/gin-gonic/gin"
+	"io"
+	"net/http"
+	"one-api/dto"
+	"one-api/relay/channel"
+	"one-api/relay/channel/openai"
+	relaycommon "one-api/relay/common"
+)
+
+type Adaptor struct {
+}
+
+func (a *Adaptor) ConvertAudioRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.AudioRequest) (io.Reader, error) {
+	//TODO implement me
+	return nil, errors.New("not implemented")
+}
+
+func (a *Adaptor) ConvertImageRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.ImageRequest) (any, error) {
+	//TODO implement me
+	return nil, errors.New("not implemented")
+}
+
+func (a *Adaptor) Init(info *relaycommon.RelayInfo) {
+}
+
+func (a *Adaptor) GetRequestURL(info *relaycommon.RelayInfo) (string, error) {
+	return fmt.Sprintf("%s/v2/chat/completions", info.BaseUrl), nil
+}
+
+func (a *Adaptor) SetupRequestHeader(c *gin.Context, req *http.Header, info *relaycommon.RelayInfo) error {
+	channel.SetupApiRequestHeader(info, c, req)
+	req.Set("Authorization", "Bearer "+info.ApiKey)
+	return nil
+}
+
+func (a *Adaptor) ConvertRequest(c *gin.Context, info *relaycommon.RelayInfo, request *dto.GeneralOpenAIRequest) (any, error) {
+	if request == nil {
+		return nil, errors.New("request is nil")
+	}
+	return request, nil
+}
+
+func (a *Adaptor) ConvertRerankRequest(c *gin.Context, relayMode int, request dto.RerankRequest) (any, error) {
+	return nil, nil
+}
+
+func (a *Adaptor) ConvertEmbeddingRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.EmbeddingRequest) (any, error) {
+	//TODO implement me
+	return nil, errors.New("not implemented")
+}
+
+func (a *Adaptor) DoRequest(c *gin.Context, info *relaycommon.RelayInfo, requestBody io.Reader) (any, error) {
+	return channel.DoApiRequest(a, c, info, requestBody)
+}
+
+func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, info *relaycommon.RelayInfo) (usage any, err *dto.OpenAIErrorWithStatusCode) {
+	if info.IsStream {
+		err, usage = openai.OaiStreamHandler(c, resp, info)
+	} else {
+		err, usage = openai.OpenaiHandler(c, resp, info.PromptTokens, info.UpstreamModelName)
+	}
+	return
+}
+
+func (a *Adaptor) GetModelList() []string {
+	return ModelList
+}
+
+func (a *Adaptor) GetChannelName() string {
+	return ChannelName
+}

relay/channel/baidu_v2/constants.go

@@ -0,0 +1,29 @@
+package baidu_v2
+
+var ModelList = []string{
+	"ernie-4.0-8k-latest",
+	"ernie-4.0-8k-preview",
+	"ernie-4.0-8k",
+	"ernie-4.0-turbo-8k-latest",
+	"ernie-4.0-turbo-8k-preview",
+	"ernie-4.0-turbo-8k",
+	"ernie-4.0-turbo-128k",
+	"ernie-3.5-8k-preview",
+	"ernie-3.5-8k",
+	"ernie-3.5-128k",
+	"ernie-speed-8k",
+	"ernie-speed-128k",
+	"ernie-speed-pro-128k",
+	"ernie-lite-8k",
+	"ernie-lite-pro-128k",
+	"ernie-tiny-8k",
+	"ernie-char-8k",
+	"ernie-char-fiction-8k",
+	"ernie-novel-8k",
+	"deepseek-v3",
+	"deepseek-r1",
+	"deepseek-r1-distill-qwen-32b",
+	"deepseek-r1-distill-qwen-14b",
+}
+
+var ChannelName = "volcengine"

relay/channel/claude/adaptor.go

@@ -9,6 +9,7 @@ import (
 	"one-api/dto"
 	"one-api/relay/channel"
 	relaycommon "one-api/relay/common"
+	"one-api/setting/model_setting"
 	"strings"
 )
 
@@ -55,6 +56,7 @@ func (a *Adaptor) SetupRequestHeader(c *gin.Context, req *http.Header, info *rel
 		anthropicVersion = "2023-06-01"
 	}
 	req.Set("anthropic-version", anthropicVersion)
+	model_setting.GetClaudeSettings().WriteHeaders(info.OriginModelName, req)
 	return nil
 }
 
@@ -73,6 +75,11 @@ func (a *Adaptor) ConvertRerankRequest(c *gin.Context, relayMode int, request dt
 	return nil, nil
 }
 
+func (a *Adaptor) ConvertEmbeddingRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.EmbeddingRequest) (any, error) {
+	//TODO implement me
+	return nil, errors.New("not implemented")
+}
+
 func (a *Adaptor) DoRequest(c *gin.Context, info *relaycommon.RelayInfo, requestBody io.Reader) (any, error) {
 	return channel.DoApiRequest(a, c, info, requestBody)
 }

relay/channel/claude/constants.go

@@ -11,6 +11,8 @@ var ModelList = []string{
 	"claude-3-5-haiku-20241022",
 	"claude-3-5-sonnet-20240620",
 	"claude-3-5-sonnet-20241022",
+	"claude-3-7-sonnet-20250219",
+	"claude-3-7-sonnet-20250219-thinking",
 }
 
 var ChannelName = "claude"

relay/channel/claude/dto.go

@@ -11,6 +11,9 @@ type ClaudeMediaMessage struct {
 	Usage       *ClaudeUsage         `json:"usage,omitempty"`
 	StopReason  *string              `json:"stop_reason,omitempty"`
 	PartialJson string               `json:"partial_json,omitempty"`
+	Thinking    string               `json:"thinking,omitempty"`
+	Signature   string               `json:"signature,omitempty"`
+	Delta       string               `json:"delta,omitempty"`
 	// tool_calls
 	Id        string `json:"id,omitempty"`
 	Name      string `json:"name,omitempty"`
@@ -54,9 +57,15 @@ type ClaudeRequest struct {
 	TopP              float64         `json:"top_p,omitempty"`
 	TopK              int             `json:"top_k,omitempty"`
 	//ClaudeMetadata    `json:"metadata,omitempty"`
-	Stream     bool   `json:"stream,omitempty"`
-	Tools      []Tool `json:"tools,omitempty"`
-	ToolChoice any    `json:"tool_choice,omitempty"`
+	Stream     bool      `json:"stream,omitempty"`
+	Tools      any       `json:"tools,omitempty"`
+	ToolChoice any       `json:"tool_choice,omitempty"`
+	Thinking   *Thinking `json:"thinking,omitempty"`
+}
+
+type Thinking struct {
+	Type         string `json:"type"`
+	BudgetTokens int    `json:"budget_tokens"`
 }
 
 type ClaudeError struct {

relay/channel/claude/relay-claude.go

@@ -1,7 +1,6 @@
 package claude
 
 import (
-	"bufio"
 	"encoding/json"
 	"fmt"
 	"io"
@@ -9,7 +8,9 @@ import (
 	"one-api/common"
 	"one-api/dto"
 	relaycommon "one-api/relay/common"
+	"one-api/relay/helper"
 	"one-api/service"
+	"one-api/setting/model_setting"
 	"strings"
 
 	"github.com/gin-gonic/gin"
@@ -92,9 +93,31 @@ func RequestOpenAI2ClaudeMessage(textRequest dto.GeneralOpenAIRequest) (*ClaudeR
 		Stream:        textRequest.Stream,
 		Tools:         claudeTools,
 	}
+
 	if claudeRequest.MaxTokens == 0 {
-		claudeRequest.MaxTokens = 4096
+		claudeRequest.MaxTokens = uint(model_setting.GetClaudeSettings().GetDefaultMaxTokens(textRequest.Model))
 	}
+
+	if model_setting.GetClaudeSettings().ThinkingAdapterEnabled &&
+		strings.HasSuffix(textRequest.Model, "-thinking") {
+
+		// 因为BudgetTokens 必须大于1024
+		if claudeRequest.MaxTokens < 1280 {
+			claudeRequest.MaxTokens = 1280
+		}
+
+		// BudgetTokens 为 max_tokens 的 80%
+		claudeRequest.Thinking = &Thinking{
+			Type:         "enabled",
+			BudgetTokens: int(float64(claudeRequest.MaxTokens) * model_setting.GetClaudeSettings().ThinkingAdapterBudgetTokensPercentage),
+		}
+		// TODO: 临时处理
+		// https://docs.anthropic.com/en/docs/build-with-claude/extended-thinking#important-considerations-when-using-extended-thinking
+		claudeRequest.TopP = 0
+		claudeRequest.Temperature = common.GetPointer[float64](1.0)
+		claudeRequest.Model = strings.TrimSuffix(textRequest.Model, "-thinking")
+	}
+
 	if textRequest.Stop != nil {
 		// stop maybe string/array string, convert to array string
 		switch textRequest.Stop.(type) {
@@ -273,7 +296,7 @@ func StreamResponseClaude2OpenAI(reqMode int, claudeResponse *ClaudeResponse) (*
 	response.Object = "chat.completion.chunk"
 	response.Model = claudeResponse.Model
 	response.Choices = make([]dto.ChatCompletionsStreamResponseChoice, 0)
-	tools := make([]dto.ToolCall, 0)
+	tools := make([]dto.ToolCallResponse, 0)
 	var choice dto.ChatCompletionsStreamResponseChoice
 	if reqMode == RequestModeCompletion {
 		choice.Delta.SetContentString(claudeResponse.Completion)
@@ -292,10 +315,10 @@ func StreamResponseClaude2OpenAI(reqMode int, claudeResponse *ClaudeResponse) (*
 			if claudeResponse.ContentBlock != nil {
 				//choice.Delta.SetContentString(claudeResponse.ContentBlock.Text)
 				if claudeResponse.ContentBlock.Type == "tool_use" {
-					tools = append(tools, dto.ToolCall{
+					tools = append(tools, dto.ToolCallResponse{
 						ID:   claudeResponse.ContentBlock.Id,
 						Type: "function",
-						Function: dto.FunctionCall{
+						Function: dto.FunctionResponse{
 							Name:      claudeResponse.ContentBlock.Name,
 							Arguments: "",
 						},
@@ -308,12 +331,20 @@ func StreamResponseClaude2OpenAI(reqMode int, claudeResponse *ClaudeResponse) (*
 			if claudeResponse.Delta != nil {
 				choice.Index = claudeResponse.Index
 				choice.Delta.SetContentString(claudeResponse.Delta.Text)
-				if claudeResponse.Delta.Type == "input_json_delta" {
-					tools = append(tools, dto.ToolCall{
-						Function: dto.FunctionCall{
+				switch claudeResponse.Delta.Type {
+				case "input_json_delta":
+					tools = append(tools, dto.ToolCallResponse{
+						Function: dto.FunctionResponse{
 							Arguments: claudeResponse.Delta.PartialJson,
 						},
 					})
+				case "signature_delta":
+					// 加密的不处理
+					signatureContent := "\n"
+					choice.Delta.ReasoningContent = &signatureContent
+				case "thinking_delta":
+					thinkingContent := claudeResponse.Delta.Thinking
+					choice.Delta.ReasoningContent = &thinkingContent
 				}
 			}
 		} else if claudeResponse.Type == "message_delta" {
@@ -351,7 +382,9 @@ func ResponseClaude2OpenAI(reqMode int, claudeResponse *ClaudeResponse) *dto.Ope
 	if len(claudeResponse.Content) > 0 {
 		responseText = claudeResponse.Content[0].Text
 	}
-	tools := make([]dto.ToolCall, 0)
+	tools := make([]dto.ToolCallResponse, 0)
+	thinkingContent := ""
+
 	if reqMode == RequestModeCompletion {
 		content, _ := json.Marshal(strings.TrimPrefix(claudeResponse.Completion, " "))
 		choice := dto.OpenAITextResponseChoice{
@@ -367,16 +400,22 @@ func ResponseClaude2OpenAI(reqMode int, claudeResponse *ClaudeResponse) *dto.Ope
 	} else {
 		fullTextResponse.Id = claudeResponse.Id
 		for _, message := range claudeResponse.Content {
-			if message.Type == "tool_use" {
+			switch message.Type {
+			case "tool_use":
 				args, _ := json.Marshal(message.Input)
-				tools = append(tools, dto.ToolCall{
+				tools = append(tools, dto.ToolCallResponse{
 					ID:   message.Id,
 					Type: "function", // compatible with other OpenAI derivative applications
-					Function: dto.FunctionCall{
+					Function: dto.FunctionResponse{
 						Name:      message.Name,
 						Arguments: string(args),
 					},
 				})
+			case "thinking":
+				// 加密的不管， 只输出明文的推理过程
+				thinkingContent = message.Thinking
+			case "text":
+				responseText = message.Text
 			}
 		}
 	}
@@ -391,6 +430,7 @@ func ResponseClaude2OpenAI(reqMode int, claudeResponse *ClaudeResponse) *dto.Ope
 	if len(tools) > 0 {
 		choice.Message.SetToolCalls(tools)
 	}
+	choice.Message.ReasoningContent = thinkingContent
 	fullTextResponse.Model = claudeResponse.Model
 	choices = append(choices, choice)
 	fullTextResponse.Choices = choices
@@ -403,28 +443,18 @@ func ClaudeStreamHandler(c *gin.Context, resp *http.Response, info *relaycommon.
 	usage = &dto.Usage{}
 	responseText := ""
 	createdTime := common.GetTimestamp()
-	scanner := bufio.NewScanner(resp.Body)
-	scanner.Split(bufio.ScanLines)
-	service.SetEventStreamHeaders(c)
 
-	for scanner.Scan() {
-		data := scanner.Text()
-		info.SetFirstResponseTime()
-		if len(data) < 6 || !strings.HasPrefix(data, "data:") {
-			continue
-		}
-		data = strings.TrimPrefix(data, "data:")
-		data = strings.TrimSpace(data)
+	helper.StreamScannerHandler(c, resp, info, func(data string) bool {
 		var claudeResponse ClaudeResponse
 		err := json.Unmarshal([]byte(data), &claudeResponse)
 		if err != nil {
 			common.SysError("error unmarshalling stream response: " + err.Error())
-			continue
+			return true
 		}
 
 		response, claudeUsage := StreamResponseClaude2OpenAI(requestMode, &claudeResponse)
 		if response == nil {
-			continue
+			return true
 		}
 		if requestMode == RequestModeCompletion {
 			responseText += claudeResponse.Completion
@@ -441,9 +471,9 @@ func ClaudeStreamHandler(c *gin.Context, resp *http.Response, info *relaycommon.
 				usage.CompletionTokens = claudeUsage.OutputTokens
 				usage.TotalTokens = claudeUsage.InputTokens + claudeUsage.OutputTokens
 			} else if claudeResponse.Type == "content_block_start" {
-
+				return true
 			} else {
-				continue
+				return true
 			}
 		}
 		//response.Id = responseId
@@ -451,11 +481,12 @@ func ClaudeStreamHandler(c *gin.Context, resp *http.Response, info *relaycommon.
 		response.Created = createdTime
 		response.Model = info.UpstreamModelName
 
-		err = service.ObjectData(c, response)
+		err = helper.ObjectData(c, response)
 		if err != nil {
 			common.LogError(c, "send_stream_response_failed: "+err.Error())
 		}
-	}
+		return true
+	})
 
 	if requestMode == RequestModeCompletion {
 		usage, _ = service.ResponseText2Usage(responseText, info.UpstreamModelName, info.PromptTokens)
@@ -468,14 +499,14 @@ func ClaudeStreamHandler(c *gin.Context, resp *http.Response, info *relaycommon.
 		}
 	}
 	if info.ShouldIncludeUsage {
-		response := service.GenerateFinalUsageResponse(responseId, createdTime, info.UpstreamModelName, *usage)
-		err := service.ObjectData(c, response)
+		response := helper.GenerateFinalUsageResponse(responseId, createdTime, info.UpstreamModelName, *usage)
+		err := helper.ObjectData(c, response)
 		if err != nil {
 			common.SysError("send final response failed: " + err.Error())
 		}
 	}
-	service.Done(c)
-	resp.Body.Close()
+	helper.Done(c)
+	//resp.Body.Close()
 	return nil, usage
 }
 

relay/channel/cloudflare/adaptor.go

@@ -4,13 +4,14 @@ import (
 	"bytes"
 	"errors"
 	"fmt"
-	"github.com/gin-gonic/gin"
 	"io"
 	"net/http"
 	"one-api/dto"
 	"one-api/relay/channel"
 	relaycommon "one-api/relay/common"
 	"one-api/relay/constant"
+
+	"github.com/gin-gonic/gin"
 )
 
 type Adaptor struct {
@@ -56,6 +57,10 @@ func (a *Adaptor) ConvertRerankRequest(c *gin.Context, relayMode int, request dt
 	return request, nil
 }
 
+func (a *Adaptor) ConvertEmbeddingRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.EmbeddingRequest) (any, error) {
+	return request, nil
+}
+
 func (a *Adaptor) ConvertAudioRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.AudioRequest) (io.Reader, error) {
 	// 添加文件字段
 	file, _, err := c.Request.FormFile("file")

relay/channel/cloudflare/relay_cloudflare.go

@@ -9,6 +9,7 @@ import (
 	"one-api/common"
 	"one-api/dto"
 	relaycommon "one-api/relay/common"
+	"one-api/relay/helper"
 	"one-api/service"
 	"strings"
 	"time"
@@ -28,8 +29,8 @@ func cfStreamHandler(c *gin.Context, resp *http.Response, info *relaycommon.Rela
 	scanner := bufio.NewScanner(resp.Body)
 	scanner.Split(bufio.ScanLines)
 
-	service.SetEventStreamHeaders(c)
-	id := service.GetResponseID(c)
+	helper.SetEventStreamHeaders(c)
+	id := helper.GetResponseID(c)
 	var responseText string
 	isFirst := true
 
@@ -57,7 +58,7 @@ func cfStreamHandler(c *gin.Context, resp *http.Response, info *relaycommon.Rela
 		}
 		response.Id = id
 		response.Model = info.UpstreamModelName
-		err = service.ObjectData(c, response)
+		err = helper.ObjectData(c, response)
 		if isFirst {
 			isFirst = false
 			info.FirstResponseTime = time.Now()
@@ -72,13 +73,13 @@ func cfStreamHandler(c *gin.Context, resp *http.Response, info *relaycommon.Rela
 	}
 	usage, _ := service.ResponseText2Usage(responseText, info.UpstreamModelName, info.PromptTokens)
 	if info.ShouldIncludeUsage {
-		response := service.GenerateFinalUsageResponse(id, info.StartTime.Unix(), info.UpstreamModelName, *usage)
-		err := service.ObjectData(c, response)
+		response := helper.GenerateFinalUsageResponse(id, info.StartTime.Unix(), info.UpstreamModelName, *usage)
+		err := helper.ObjectData(c, response)
 		if err != nil {
 			common.LogError(c, "error_rendering_final_usage_response: "+err.Error())
 		}
 	}
-	service.Done(c)
+	helper.Done(c)
 
 	err := resp.Body.Close()
 	if err != nil {
@@ -109,7 +110,7 @@ func cfHandler(c *gin.Context, resp *http.Response, info *relaycommon.RelayInfo)
 	}
 	usage, _ := service.ResponseText2Usage(responseText, info.UpstreamModelName, info.PromptTokens)
 	response.Usage = *usage
-	response.Id = service.GetResponseID(c)
+	response.Id = helper.GetResponseID(c)
 	jsonResponse, err := json.Marshal(response)
 	if err != nil {
 		return service.OpenAIErrorWrapper(err, "marshal_response_body_failed", http.StatusInternalServerError), nil

relay/channel/cohere/adaptor.go

@@ -54,6 +54,12 @@ func (a *Adaptor) ConvertRerankRequest(c *gin.Context, relayMode int, request dt
 	return requestConvertRerank2Cohere(request), nil
 }
 
+func (a *Adaptor) ConvertEmbeddingRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.EmbeddingRequest) (any, error) {
+	//TODO implement me
+	return nil, errors.New("not implemented")
+}
+
+
 func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, info *relaycommon.RelayInfo) (usage any, err *dto.OpenAIErrorWithStatusCode) {
 	if info.RelayMode == constant.RelayModeRerank {
 		err, usage = cohereRerankHandler(c, resp, info)

relay/channel/cohere/relay-cohere.go

@@ -10,6 +10,7 @@ import (
 	"one-api/common"
 	"one-api/dto"
 	relaycommon "one-api/relay/common"
+	"one-api/relay/helper"
 	"one-api/service"
 	"strings"
 	"time"
@@ -103,7 +104,7 @@ func cohereStreamHandler(c *gin.Context, resp *http.Response, info *relaycommon.
 		}
 		stopChan <- true
 	}()
-	service.SetEventStreamHeaders(c)
+	helper.SetEventStreamHeaders(c)
 	isFirst := true
 	c.Stream(func(w io.Writer) bool {
 		select {

relay/channel/deepseek/adaptor.go

@@ -10,6 +10,7 @@ import (
 	"one-api/relay/channel"
 	"one-api/relay/channel/openai"
 	relaycommon "one-api/relay/common"
+	"one-api/relay/constant"
 )
 
 type Adaptor struct {
@@ -29,7 +30,12 @@ func (a *Adaptor) Init(info *relaycommon.RelayInfo) {
 }
 
 func (a *Adaptor) GetRequestURL(info *relaycommon.RelayInfo) (string, error) {
-	return fmt.Sprintf("%s/chat/completions", info.BaseUrl), nil
+	switch info.RelayMode {
+	case constant.RelayModeCompletions:
+		return fmt.Sprintf("%s/beta/completions", info.BaseUrl), nil
+	default:
+		return fmt.Sprintf("%s/v1/chat/completions", info.BaseUrl), nil
+	}
 }
 
 func (a *Adaptor) SetupRequestHeader(c *gin.Context, req *http.Header, info *relaycommon.RelayInfo) error {
@@ -49,6 +55,11 @@ func (a *Adaptor) ConvertRerankRequest(c *gin.Context, relayMode int, request dt
 	return nil, nil
 }
 
+func (a *Adaptor) ConvertEmbeddingRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.EmbeddingRequest) (any, error) {
+	//TODO implement me
+	return nil, errors.New("not implemented")
+}
+
 func (a *Adaptor) DoRequest(c *gin.Context, info *relaycommon.RelayInfo, requestBody io.Reader) (any, error) {
 	return channel.DoApiRequest(a, c, info, requestBody)
 }

relay/channel/dify/adaptor.go

@@ -9,9 +9,18 @@ import (
 	"one-api/dto"
 	"one-api/relay/channel"
 	relaycommon "one-api/relay/common"
+	"strings"
+)
+
+const (
+	BotTypeChatFlow   = 1 // chatflow default
+	BotTypeAgent      = 2
+	BotTypeWorkFlow   = 3
+	BotTypeCompletion = 4
 )
 
 type Adaptor struct {
+	BotType int
 }
 
 func (a *Adaptor) ConvertAudioRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.AudioRequest) (io.Reader, error) {
@@ -25,10 +34,28 @@ func (a *Adaptor) ConvertImageRequest(c *gin.Context, info *relaycommon.RelayInf
 }
 
 func (a *Adaptor) Init(info *relaycommon.RelayInfo) {
+	if strings.HasPrefix(info.UpstreamModelName, "agent") {
+		a.BotType = BotTypeAgent
+	} else if strings.HasPrefix(info.UpstreamModelName, "workflow") {
+		a.BotType = BotTypeWorkFlow
+	} else if strings.HasPrefix(info.UpstreamModelName, "chat") {
+		a.BotType = BotTypeCompletion
+	} else {
+		a.BotType = BotTypeChatFlow
+	}
 }
 
 func (a *Adaptor) GetRequestURL(info *relaycommon.RelayInfo) (string, error) {
-	return fmt.Sprintf("%s/v1/chat-messages", info.BaseUrl), nil
+	switch a.BotType {
+	case BotTypeWorkFlow:
+		return fmt.Sprintf("%s/v1/workflows/run", info.BaseUrl), nil
+	case BotTypeCompletion:
+		return fmt.Sprintf("%s/v1/completion-messages", info.BaseUrl), nil
+	case BotTypeAgent:
+		fallthrough
+	default:
+		return fmt.Sprintf("%s/v1/chat-messages", info.BaseUrl), nil
+	}
 }
 
 func (a *Adaptor) SetupRequestHeader(c *gin.Context, req *http.Header, info *relaycommon.RelayInfo) error {
@@ -48,6 +75,11 @@ func (a *Adaptor) ConvertRerankRequest(c *gin.Context, relayMode int, request dt
 	return nil, nil
 }
 
+func (a *Adaptor) ConvertEmbeddingRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.EmbeddingRequest) (any, error) {
+	//TODO implement me
+	return nil, errors.New("not implemented")
+}
+
 func (a *Adaptor) DoRequest(c *gin.Context, info *relaycommon.RelayInfo, requestBody io.Reader) (any, error) {
 	return channel.DoApiRequest(a, c, info, requestBody)
 }

relay/channel/dify/relay-dify.go

@@ -10,6 +10,7 @@ import (
 	"one-api/constant"
 	"one-api/dto"
 	relaycommon "one-api/relay/common"
+	"one-api/relay/helper"
 	"one-api/service"
 	"strings"
 )
@@ -66,7 +67,7 @@ func difyStreamHandler(c *gin.Context, resp *http.Response, info *relaycommon.Re
 	scanner := bufio.NewScanner(resp.Body)
 	scanner.Split(bufio.ScanLines)
 
-	service.SetEventStreamHeaders(c)
+	helper.SetEventStreamHeaders(c)
 
 	for scanner.Scan() {
 		data := scanner.Text()
@@ -92,7 +93,7 @@ func difyStreamHandler(c *gin.Context, resp *http.Response, info *relaycommon.Re
 				responseText += openaiResponse.Choices[0].Delta.GetContentString()
 			}
 		}
-		err = service.ObjectData(c, openaiResponse)
+		err = helper.ObjectData(c, openaiResponse)
 		if err != nil {
 			common.SysError(err.Error())
 		}
@@ -100,7 +101,7 @@ func difyStreamHandler(c *gin.Context, resp *http.Response, info *relaycommon.Re
 	if err := scanner.Err(); err != nil {
 		common.SysError("error reading stream: " + err.Error())
 	}
-	service.Done(c)
+	helper.Done(c)
 	err := resp.Body.Close()
 	if err != nil {
 		//return service.OpenAIErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), nil

relay/channel/gemini/adaptor.go

@@ -1,15 +1,21 @@
 package gemini
 
 import (
+	"encoding/json"
 	"errors"
 	"fmt"
-	"github.com/gin-gonic/gin"
 	"io"
 	"net/http"
-	"one-api/constant"
+	"one-api/common"
 	"one-api/dto"
 	"one-api/relay/channel"
 	relaycommon "one-api/relay/common"
+	"one-api/service"
+	"one-api/setting/model_setting"
+
+	"strings"
+
+	"github.com/gin-gonic/gin"
 )
 
 type Adaptor struct {
@@ -21,8 +27,36 @@ func (a *Adaptor) ConvertAudioRequest(c *gin.Context, info *relaycommon.RelayInf
 }
 
 func (a *Adaptor) ConvertImageRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.ImageRequest) (any, error) {
-	//TODO implement me
-	return nil, errors.New("not implemented")
+	if !strings.HasPrefix(info.UpstreamModelName, "imagen") {
+		return nil, errors.New("not supported model for image generation")
+	}
+
+	// convert size to aspect ratio
+	aspectRatio := "1:1" // default aspect ratio
+	switch request.Size {
+	case "1024x1024":
+		aspectRatio = "1:1"
+	case "1024x1792":
+		aspectRatio = "9:16"
+	case "1792x1024":
+		aspectRatio = "16:9"
+	}
+
+	// build gemini imagen request
+	geminiRequest := GeminiImageRequest{
+		Instances: []GeminiImageInstance{
+			{
+				Prompt: request.Prompt,
+			},
+		},
+		Parameters: GeminiImageParameters{
+			SampleCount:      request.N,
+			AspectRatio:      aspectRatio,
+			PersonGeneration: "allow_adult", // default allow adult
+		},
+	}
+
+	return geminiRequest, nil
 }
 
 func (a *Adaptor) Init(info *relaycommon.RelayInfo) {
@@ -30,14 +64,16 @@ func (a *Adaptor) Init(info *relaycommon.RelayInfo) {
 }
 
 func (a *Adaptor) GetRequestURL(info *relaycommon.RelayInfo) (string, error) {
-	// 从映射中获取模型名称对应的版本，如果找不到就使用 info.ApiVersion 或默认的版本 "v1beta"
-	version, beta := constant.GeminiModelMap[info.UpstreamModelName]
-	if !beta {
-		if info.ApiVersion != "" {
-			version = info.ApiVersion
-		} else {
-			version = "v1beta"
-		}
+	version := model_setting.GetGeminiVersionSetting(info.UpstreamModelName)
+
+	if strings.HasPrefix(info.UpstreamModelName, "imagen") {
+		return fmt.Sprintf("%s/%s/models/%s:predict", info.BaseUrl, version, info.UpstreamModelName), nil
+	}
+
+	if strings.HasPrefix(info.UpstreamModelName, "text-embedding") ||
+		strings.HasPrefix(info.UpstreamModelName, "embedding") ||
+		strings.HasPrefix(info.UpstreamModelName, "gemini-embedding") {
+		return fmt.Sprintf("%s/%s/models/%s:embedContent", info.BaseUrl, version, info.UpstreamModelName), nil
 	}
 
 	action := "generateContent"
@@ -68,11 +104,56 @@ func (a *Adaptor) ConvertRerankRequest(c *gin.Context, relayMode int, request dt
 	return nil, nil
 }
 
+func (a *Adaptor) ConvertEmbeddingRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.EmbeddingRequest) (any, error) {
+	if request.Input == nil {
+		return nil, errors.New("input is required")
+	}
+
+	inputs := request.ParseInput()
+	if len(inputs) == 0 {
+		return nil, errors.New("input is empty")
+	}
+
+	// only process the first input
+	geminiRequest := GeminiEmbeddingRequest{
+		Content: GeminiChatContent{
+			Parts: []GeminiPart{
+				{
+					Text: inputs[0],
+				},
+			},
+		},
+	}
+
+	// set specific parameters for different models
+	// https://ai.google.dev/api/embeddings?hl=zh-cn#method:-models.embedcontent
+	switch info.UpstreamModelName {
+	case "text-embedding-004":
+		// except embedding-001 supports setting `OutputDimensionality`
+		if request.Dimensions > 0 {
+			geminiRequest.OutputDimensionality = request.Dimensions
+		}
+	}
+
+	return geminiRequest, nil
+}
+
 func (a *Adaptor) DoRequest(c *gin.Context, info *relaycommon.RelayInfo, requestBody io.Reader) (any, error) {
 	return channel.DoApiRequest(a, c, info, requestBody)
 }
 
 func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, info *relaycommon.RelayInfo) (usage any, err *dto.OpenAIErrorWithStatusCode) {
+	if strings.HasPrefix(info.UpstreamModelName, "imagen") {
+		return GeminiImageHandler(c, resp, info)
+	}
+
+	// check if the model is an embedding model
+	if strings.HasPrefix(info.UpstreamModelName, "text-embedding") ||
+		strings.HasPrefix(info.UpstreamModelName, "embedding") ||
+		strings.HasPrefix(info.UpstreamModelName, "gemini-embedding") {
+		return GeminiEmbeddingHandler(c, resp, info)
+	}
+
 	if info.IsStream {
 		err, usage = GeminiChatStreamHandler(c, resp, info)
 	} else {
@@ -81,6 +162,60 @@ func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, info *relaycom
 	return
 }
 
+func GeminiImageHandler(c *gin.Context, resp *http.Response, info *relaycommon.RelayInfo) (usage any, err *dto.OpenAIErrorWithStatusCode) {
+	responseBody, readErr := io.ReadAll(resp.Body)
+	if readErr != nil {
+		return nil, service.OpenAIErrorWrapper(readErr, "read_response_body_failed", http.StatusInternalServerError)
+	}
+	_ = resp.Body.Close()
+
+	var geminiResponse GeminiImageResponse
+	if jsonErr := json.Unmarshal(responseBody, &geminiResponse); jsonErr != nil {
+		return nil, service.OpenAIErrorWrapper(jsonErr, "unmarshal_response_body_failed", http.StatusInternalServerError)
+	}
+
+	if len(geminiResponse.Predictions) == 0 {
+		return nil, service.OpenAIErrorWrapper(errors.New("no images generated"), "no_images", http.StatusBadRequest)
+	}
+
+	// convert to openai format response
+	openAIResponse := dto.ImageResponse{
+		Created: common.GetTimestamp(),
+		Data:    make([]dto.ImageData, 0, len(geminiResponse.Predictions)),
+	}
+
+	for _, prediction := range geminiResponse.Predictions {
+		if prediction.RaiFilteredReason != "" {
+			continue // skip filtered image
+		}
+		openAIResponse.Data = append(openAIResponse.Data, dto.ImageData{
+			B64Json: prediction.BytesBase64Encoded,
+		})
+	}
+
+	jsonResponse, jsonErr := json.Marshal(openAIResponse)
+	if jsonErr != nil {
+		return nil, service.OpenAIErrorWrapper(jsonErr, "marshal_response_failed", http.StatusInternalServerError)
+	}
+
+	c.Writer.Header().Set("Content-Type", "application/json")
+	c.Writer.WriteHeader(resp.StatusCode)
+	_, _ = c.Writer.Write(jsonResponse)
+
+	// https://github.com/google-gemini/cookbook/blob/719a27d752aac33f39de18a8d3cb42a70874917e/quickstarts/Counting_Tokens.ipynb
+	// each image has fixed 258 tokens
+	const imageTokens = 258
+	generatedImages := len(openAIResponse.Data)
+
+	usage = &dto.Usage{
+		PromptTokens:     imageTokens * generatedImages, // each generated image has fixed 258 tokens
+		CompletionTokens: 0,                             // image generation does not calculate completion tokens
+		TotalTokens:      imageTokens * generatedImages,
+	}
+
+	return usage, nil
+}
+
 func (a *Adaptor) GetModelList() []string {
 	return ModelList
 }

relay/channel/gemini/constant.go

@@ -3,17 +3,33 @@ package gemini
 var ModelList = []string{
 	// stable version
 	"gemini-1.5-pro", "gemini-1.5-flash", "gemini-1.5-flash-8b",
+	"gemini-2.0-flash",
 	// latest version
 	"gemini-1.5-pro-latest", "gemini-1.5-flash-latest",
-	// legacy version
-	"gemini-1.5-pro-exp-0827", "gemini-1.5-flash-exp-0827",
-	// exp
-	"gemini-exp-1114", "gemini-exp-1121", "gemini-exp-1206",
+	// preview version
+	"gemini-2.0-flash-lite-preview",
+	// gemini exp
+	"gemini-exp-1206",
 	// flash exp
 	"gemini-2.0-flash-exp",
+	// pro exp
+	"gemini-2.0-pro-exp",
 	// thinking exp
 	"gemini-2.0-flash-thinking-exp",
-	"gemini-2.0-flash-thinking-exp-1219",
+	// imagen models
+	"imagen-3.0-generate-002",
+	// embedding models
+	"gemini-embedding-exp-03-07",
+	"text-embedding-004",
+	"embedding-001",
+}
+
+var SafetySettingList = []string{
+	"HARM_CATEGORY_HARASSMENT",
+	"HARM_CATEGORY_HATE_SPEECH",
+	"HARM_CATEGORY_SEXUALLY_EXPLICIT",
+	"HARM_CATEGORY_DANGEROUS_CONTENT",
+	"HARM_CATEGORY_CIVIC_INTEGRITY",
 }
 
 var ChannelName = "google gemini"

relay/channel/gemini/dto.go

@@ -109,3 +109,46 @@ type GeminiUsageMetadata struct {
 	CandidatesTokenCount int `json:"candidatesTokenCount"`
 	TotalTokenCount      int `json:"totalTokenCount"`
 }
+
+// Imagen related structs
+type GeminiImageRequest struct {
+	Instances  []GeminiImageInstance `json:"instances"`
+	Parameters GeminiImageParameters `json:"parameters"`
+}
+
+type GeminiImageInstance struct {
+	Prompt string `json:"prompt"`
+}
+
+type GeminiImageParameters struct {
+	SampleCount      int    `json:"sampleCount,omitempty"`
+	AspectRatio      string `json:"aspectRatio,omitempty"`
+	PersonGeneration string `json:"personGeneration,omitempty"`
+}
+
+type GeminiImageResponse struct {
+	Predictions []GeminiImagePrediction `json:"predictions"`
+}
+
+type GeminiImagePrediction struct {
+	MimeType           string `json:"mimeType"`
+	BytesBase64Encoded string `json:"bytesBase64Encoded"`
+	RaiFilteredReason  string `json:"raiFilteredReason,omitempty"`
+	SafetyAttributes   any    `json:"safetyAttributes,omitempty"`
+}
+
+// Embedding related structs
+type GeminiEmbeddingRequest struct {
+	Content              GeminiChatContent `json:"content"`
+	TaskType             string            `json:"taskType,omitempty"`
+	Title                string            `json:"title,omitempty"`
+	OutputDimensionality int               `json:"outputDimensionality,omitempty"`
+}
+
+type GeminiEmbeddingResponse struct {
+	Embedding ContentEmbedding `json:"embedding"`
+}
+
+type ContentEmbedding struct {
+	Values []float64 `json:"values"`
+}

relay/channel/gemini/relay-gemini.go

@@ -1,7 +1,6 @@
 package gemini
 
 import (
-	"bufio"
 	"encoding/json"
 	"fmt"
 	"io"
@@ -10,7 +9,9 @@ import (
 	"one-api/constant"
 	"one-api/dto"
 	relaycommon "one-api/relay/common"
+	"one-api/relay/helper"
 	"one-api/service"
+	"one-api/setting/model_setting"
 	"strings"
 	"unicode/utf8"
 
@@ -22,28 +23,7 @@ func CovertGemini2OpenAI(textRequest dto.GeneralOpenAIRequest) (*GeminiChatReque
 
 	geminiRequest := GeminiChatRequest{
 		Contents: make([]GeminiChatContent, 0, len(textRequest.Messages)),
-		SafetySettings: []GeminiChatSafetySettings{
-			{
-				Category:  "HARM_CATEGORY_HARASSMENT",
-				Threshold: common.GeminiSafetySetting,
-			},
-			{
-				Category:  "HARM_CATEGORY_HATE_SPEECH",
-				Threshold: common.GeminiSafetySetting,
-			},
-			{
-				Category:  "HARM_CATEGORY_SEXUALLY_EXPLICIT",
-				Threshold: common.GeminiSafetySetting,
-			},
-			{
-				Category:  "HARM_CATEGORY_DANGEROUS_CONTENT",
-				Threshold: common.GeminiSafetySetting,
-			},
-			{
-				Category:  "HARM_CATEGORY_CIVIC_INTEGRITY",
-				Threshold: common.GeminiSafetySetting,
-			},
-		},
+		//SafetySettings: []GeminiChatSafetySettings{},
 		GenerationConfig: GeminiChatGenerationConfig{
 			Temperature:     textRequest.Temperature,
 			TopP:            textRequest.TopP,
@@ -52,9 +32,18 @@ func CovertGemini2OpenAI(textRequest dto.GeneralOpenAIRequest) (*GeminiChatReque
 		},
 	}
 
+	safetySettings := make([]GeminiChatSafetySettings, 0, len(SafetySettingList))
+	for _, category := range SafetySettingList {
+		safetySettings = append(safetySettings, GeminiChatSafetySettings{
+			Category:  category,
+			Threshold: model_setting.GetGeminiSafetySetting(category),
+		})
+	}
+	geminiRequest.SafetySettings = safetySettings
+
 	// openaiContent.FuncToToolCalls()
 	if textRequest.Tools != nil {
-		functions := make([]dto.FunctionCall, 0, len(textRequest.Tools))
+		functions := make([]dto.FunctionRequest, 0, len(textRequest.Tools))
 		googleSearch := false
 		codeExecution := false
 		for _, tool := range textRequest.Tools {
@@ -349,7 +338,7 @@ func unescapeMapOrSlice(data interface{}) interface{} {
 	return data
 }
 
-func getToolCall(item *GeminiPart) *dto.ToolCall {
+func getResponseToolCall(item *GeminiPart) *dto.ToolCallResponse {
 	var argsBytes []byte
 	var err error
 	if result, ok := item.FunctionCall.Arguments.(map[string]interface{}); ok {
@@ -361,10 +350,10 @@ func getToolCall(item *GeminiPart) *dto.ToolCall {
 	if err != nil {
 		return nil
 	}
-	return &dto.ToolCall{
+	return &dto.ToolCallResponse{
 		ID:   fmt.Sprintf("call_%s", common.GetUUID()),
 		Type: "function",
-		Function: dto.FunctionCall{
+		Function: dto.FunctionResponse{
 			Arguments: string(argsBytes),
 			Name:      item.FunctionCall.FunctionName,
 		},
@@ -379,7 +368,7 @@ func responseGeminiChat2OpenAI(response *GeminiChatResponse) *dto.OpenAITextResp
 		Choices: make([]dto.OpenAITextResponseChoice, 0, len(response.Candidates)),
 	}
 	content, _ := json.Marshal("")
-	is_tool_call := false
+	isToolCall := false
 	for _, candidate := range response.Candidates {
 		choice := dto.OpenAITextResponseChoice{
 			Index: int(candidate.Index),
@@ -391,12 +380,12 @@ func responseGeminiChat2OpenAI(response *GeminiChatResponse) *dto.OpenAITextResp
 		}
 		if len(candidate.Content.Parts) > 0 {
 			var texts []string
-			var tool_calls []dto.ToolCall
+			var toolCalls []dto.ToolCallResponse
 			for _, part := range candidate.Content.Parts {
 				if part.FunctionCall != nil {
 					choice.FinishReason = constant.FinishReasonToolCalls
-					if call := getToolCall(&part); call != nil {
-						tool_calls = append(tool_calls, *call)
+					if call := getResponseToolCall(&part); call != nil {
+						toolCalls = append(toolCalls, *call)
 					}
 				} else {
 					if part.ExecutableCode != nil {
@@ -411,9 +400,9 @@ func responseGeminiChat2OpenAI(response *GeminiChatResponse) *dto.OpenAITextResp
 					}
 				}
 			}
-			if len(tool_calls) > 0 {
-				choice.Message.SetToolCalls(tool_calls)
-				is_tool_call = true
+			if len(toolCalls) > 0 {
+				choice.Message.SetToolCalls(toolCalls)
+				isToolCall = true
 			}
 
 			choice.Message.SetStringContent(strings.Join(texts, "\n"))
@@ -429,7 +418,7 @@ func responseGeminiChat2OpenAI(response *GeminiChatResponse) *dto.OpenAITextResp
 				choice.FinishReason = constant.FinishReasonContentFilter
 			}
 		}
-		if is_tool_call {
+		if isToolCall {
 			choice.FinishReason = constant.FinishReasonToolCalls
 		}
 
@@ -440,10 +429,10 @@ func responseGeminiChat2OpenAI(response *GeminiChatResponse) *dto.OpenAITextResp
 
 func streamResponseGeminiChat2OpenAI(geminiResponse *GeminiChatResponse) (*dto.ChatCompletionsStreamResponse, bool) {
 	choices := make([]dto.ChatCompletionsStreamResponseChoice, 0, len(geminiResponse.Candidates))
-	is_stop := false
+	isStop := false
 	for _, candidate := range geminiResponse.Candidates {
 		if candidate.FinishReason != nil && *candidate.FinishReason == "STOP" {
-			is_stop = true
+			isStop = true
 			candidate.FinishReason = nil
 		}
 		choice := dto.ChatCompletionsStreamResponseChoice{
@@ -468,7 +457,7 @@ func streamResponseGeminiChat2OpenAI(geminiResponse *GeminiChatResponse) (*dto.C
 		for _, part := range candidate.Content.Parts {
 			if part.FunctionCall != nil {
 				isTools = true
-				if call := getToolCall(&part); call != nil {
+				if call := getResponseToolCall(&part); call != nil {
 					call.SetIndex(len(choice.Delta.ToolCalls))
 					choice.Delta.ToolCalls = append(choice.Delta.ToolCalls, *call)
 				}
@@ -493,9 +482,8 @@ func streamResponseGeminiChat2OpenAI(geminiResponse *GeminiChatResponse) (*dto.C
 
 	var response dto.ChatCompletionsStreamResponse
 	response.Object = "chat.completion.chunk"
-	response.Model = "gemini"
 	response.Choices = choices
-	return &response, is_stop
+	return &response, isStop
 }
 
 func GeminiChatStreamHandler(c *gin.Context, resp *http.Response, info *relaycommon.RelayInfo) (*dto.OpenAIErrorWithStatusCode, *dto.Usage) {
@@ -503,27 +491,16 @@ func GeminiChatStreamHandler(c *gin.Context, resp *http.Response, info *relaycom
 	id := fmt.Sprintf("chatcmpl-%s", common.GetUUID())
 	createAt := common.GetTimestamp()
 	var usage = &dto.Usage{}
-	scanner := bufio.NewScanner(resp.Body)
-	scanner.Split(bufio.ScanLines)
 
-	service.SetEventStreamHeaders(c)
-	for scanner.Scan() {
-		data := scanner.Text()
-		info.SetFirstResponseTime()
-		data = strings.TrimSpace(data)
-		if !strings.HasPrefix(data, "data: ") {
-			continue
-		}
-		data = strings.TrimPrefix(data, "data: ")
-		data = strings.TrimSuffix(data, "\"")
+	helper.StreamScannerHandler(c, resp, info, func(data string) bool {
 		var geminiResponse GeminiChatResponse
 		err := json.Unmarshal([]byte(data), &geminiResponse)
 		if err != nil {
 			common.LogError(c, "error unmarshalling stream response: "+err.Error())
-			continue
+			return false
 		}
 
-		response, is_stop := streamResponseGeminiChat2OpenAI(&geminiResponse)
+		response, isStop := streamResponseGeminiChat2OpenAI(&geminiResponse)
 		response.Id = id
 		response.Created = createAt
 		response.Model = info.UpstreamModelName
@@ -532,15 +509,16 @@ func GeminiChatStreamHandler(c *gin.Context, resp *http.Response, info *relaycom
 			usage.PromptTokens = geminiResponse.UsageMetadata.PromptTokenCount
 			usage.CompletionTokens = geminiResponse.UsageMetadata.CandidatesTokenCount
 		}
-		err = service.ObjectData(c, response)
+		err = helper.ObjectData(c, response)
 		if err != nil {
 			common.LogError(c, err.Error())
 		}
-		if is_stop {
-			response := service.GenerateStopResponse(id, createAt, info.UpstreamModelName, constant.FinishReasonStop)
-			service.ObjectData(c, response)
+		if isStop {
+			response := helper.GenerateStopResponse(id, createAt, info.UpstreamModelName, constant.FinishReasonStop)
+			helper.ObjectData(c, response)
 		}
-	}
+		return true
+	})
 
 	var response *dto.ChatCompletionsStreamResponse
 
@@ -549,14 +527,14 @@ func GeminiChatStreamHandler(c *gin.Context, resp *http.Response, info *relaycom
 	usage.CompletionTokenDetails.TextTokens = usage.CompletionTokens
 
 	if info.ShouldIncludeUsage {
-		response = service.GenerateFinalUsageResponse(id, createAt, info.UpstreamModelName, *usage)
-		err := service.ObjectData(c, response)
+		response = helper.GenerateFinalUsageResponse(id, createAt, info.UpstreamModelName, *usage)
+		err := helper.ObjectData(c, response)
 		if err != nil {
 			common.SysError("send final response failed: " + err.Error())
 		}
 	}
-	service.Done(c)
-	resp.Body.Close()
+	helper.Done(c)
+	//resp.Body.Close()
 	return nil, usage
 }
 
@@ -602,3 +580,52 @@ func GeminiChatHandler(c *gin.Context, resp *http.Response, info *relaycommon.Re
 	_, err = c.Writer.Write(jsonResponse)
 	return nil, &usage
 }
+
+func GeminiEmbeddingHandler(c *gin.Context, resp *http.Response, info *relaycommon.RelayInfo) (usage any, err *dto.OpenAIErrorWithStatusCode) {
+	responseBody, readErr := io.ReadAll(resp.Body)
+	if readErr != nil {
+		return nil, service.OpenAIErrorWrapper(readErr, "read_response_body_failed", http.StatusInternalServerError)
+	}
+	_ = resp.Body.Close()
+
+	var geminiResponse GeminiEmbeddingResponse
+	if jsonErr := json.Unmarshal(responseBody, &geminiResponse); jsonErr != nil {
+		return nil, service.OpenAIErrorWrapper(jsonErr, "unmarshal_response_body_failed", http.StatusInternalServerError)
+	}
+
+	// convert to openai format response
+	openAIResponse := dto.OpenAIEmbeddingResponse{
+		Object: "list",
+		Data: []dto.OpenAIEmbeddingResponseItem{
+			{
+				Object:    "embedding",
+				Embedding: geminiResponse.Embedding.Values,
+				Index:     0,
+			},
+		},
+		Model: info.UpstreamModelName,
+	}
+
+	// calculate usage
+	// https://ai.google.dev/gemini-api/docs/pricing?hl=zh-cn#text-embedding-004
+	// Google has not yet clarified how embedding models will be billed
+	// refer to openai billing method to use input tokens billing
+	// https://platform.openai.com/docs/guides/embeddings#what-are-embeddings
+	usage = &dto.Usage{
+		PromptTokens:     info.PromptTokens,
+		CompletionTokens: 0,
+		TotalTokens:      info.PromptTokens,
+	}
+	openAIResponse.Usage = *usage.(*dto.Usage)
+
+	jsonResponse, jsonErr := json.Marshal(openAIResponse)
+	if jsonErr != nil {
+		return nil, service.OpenAIErrorWrapper(jsonErr, "marshal_response_failed", http.StatusInternalServerError)
+	}
+
+	c.Writer.Header().Set("Content-Type", "application/json")
+	c.Writer.WriteHeader(resp.StatusCode)
+	_, _ = c.Writer.Write(jsonResponse)
+
+	return usage, nil
+}

relay/channel/jina/adaptor.go

@@ -55,9 +55,13 @@ func (a *Adaptor) ConvertRerankRequest(c *gin.Context, relayMode int, request dt
 	return request, nil
 }
 
+func (a *Adaptor) ConvertEmbeddingRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.EmbeddingRequest) (any, error) {
+	return request, nil
+}
+
 func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, info *relaycommon.RelayInfo) (usage any, err *dto.OpenAIErrorWithStatusCode) {
 	if info.RelayMode == constant.RelayModeRerank {
-		err, usage = jinaRerankHandler(c, resp)
+		err, usage = JinaRerankHandler(c, resp)
 	} else if info.RelayMode == constant.RelayModeEmbeddings {
 		err, usage = jinaEmbeddingHandler(c, resp)
 	}

relay/channel/jina/relay-jina.go

@@ -9,7 +9,7 @@ import (
 	"one-api/service"
 )
 
-func jinaRerankHandler(c *gin.Context, resp *http.Response) (*dto.OpenAIErrorWithStatusCode, *dto.Usage) {
+func JinaRerankHandler(c *gin.Context, resp *http.Response) (*dto.OpenAIErrorWithStatusCode, *dto.Usage) {
 	responseBody, err := io.ReadAll(resp.Body)
 	if err != nil {
 		return service.OpenAIErrorWrapper(err, "read_response_body_failed", http.StatusInternalServerError), nil

relay/channel/mistral/adaptor.go

@@ -41,15 +41,18 @@ func (a *Adaptor) ConvertRequest(c *gin.Context, info *relaycommon.RelayInfo, re
 	if request == nil {
 		return nil, errors.New("request is nil")
 	}
-	mistralReq := requestOpenAI2Mistral(*request)
-	//common.LogJson(c, "body", mistralReq)
-	return mistralReq, nil
+	return requestOpenAI2Mistral(request), nil
 }
 
 func (a *Adaptor) ConvertRerankRequest(c *gin.Context, relayMode int, request dto.RerankRequest) (any, error) {
 	return nil, nil
 }
 
+func (a *Adaptor) ConvertEmbeddingRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.EmbeddingRequest) (any, error) {
+	//TODO implement me
+	return nil, errors.New("not implemented")
+}
+
 func (a *Adaptor) DoRequest(c *gin.Context, info *relaycommon.RelayInfo, requestBody io.Reader) (any, error) {
 	return channel.DoApiRequest(a, c, info, requestBody)
 }

relay/channel/mistral/text.go

@@ -1,25 +1,21 @@
 package mistral
 
 import (
-	"encoding/json"
 	"one-api/dto"
 )
 
-func requestOpenAI2Mistral(request dto.GeneralOpenAIRequest) *dto.GeneralOpenAIRequest {
+func requestOpenAI2Mistral(request *dto.GeneralOpenAIRequest) *dto.GeneralOpenAIRequest {
 	messages := make([]dto.Message, 0, len(request.Messages))
 	for _, message := range request.Messages {
-		if !message.IsStringContent() {
-			mediaMessages := message.ParseContent()
-			for j, mediaMessage := range mediaMessages {
-				if mediaMessage.Type == dto.ContentTypeImageURL {
-					imageUrl := mediaMessage.ImageUrl.(dto.MessageImageUrl)
-					mediaMessage.ImageUrl = imageUrl.Url
-					mediaMessages[j] = mediaMessage
-				}
+		mediaMessages := message.ParseContent()
+		for j, mediaMessage := range mediaMessages {
+			if mediaMessage.Type == dto.ContentTypeImageURL {
+				imageUrl := mediaMessage.ImageUrl.(dto.MessageImageUrl)
+				mediaMessage.ImageUrl = imageUrl.Url
+				mediaMessages[j] = mediaMessage
 			}
-			messageRaw, _ := json.Marshal(mediaMessages)
-			message.Content = messageRaw
 		}
+		message.SetMediaContent(mediaMessages)
 		messages = append(messages, dto.Message{
 			Role:       message.Role,
 			Content:    message.Content,

relay/channel/mokaai/adaptor.go

@@ -0,0 +1,93 @@
+package mokaai
+
+import (
+	"errors"
+	"fmt"
+	"github.com/gin-gonic/gin"
+	"io"
+	"net/http"
+	"one-api/dto"
+	"one-api/relay/channel"
+	relaycommon "one-api/relay/common"
+	"one-api/relay/constant"
+	"strings"
+)
+
+type Adaptor struct {
+}
+
+func (a *Adaptor) ConvertAudioRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.AudioRequest) (io.Reader, error) {
+	//TODO implement me
+	return nil, errors.New("not implemented")
+}
+
+func (a *Adaptor) ConvertImageRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.ImageRequest) (any, error) {
+	//TODO implement me
+	return nil, errors.New("not implemented")
+}
+
+func (a *Adaptor) ConvertEmbeddingRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.EmbeddingRequest) (any, error) {
+	//TODO implement me
+	return request, nil
+}
+
+func (a *Adaptor) Init(info *relaycommon.RelayInfo) {
+
+}
+
+func (a *Adaptor) GetRequestURL(info *relaycommon.RelayInfo) (string, error) {
+	// https://cloud.baidu.com/doc/WENXINWORKSHOP/s/clntwmv7t
+	suffix := "chat/"
+	if strings.HasPrefix(info.UpstreamModelName, "m3e") {
+		suffix = "embeddings"
+	}
+	fullRequestURL := fmt.Sprintf("%s/%s", info.BaseUrl, suffix)
+	return fullRequestURL, nil
+}
+
+func (a *Adaptor) SetupRequestHeader(c *gin.Context, req *http.Header, info *relaycommon.RelayInfo) error {
+	channel.SetupApiRequestHeader(info, c, req)
+	req.Set("Authorization", fmt.Sprintf("Bearer %s", info.ApiKey))
+	return nil
+}
+
+func (a *Adaptor) ConvertRequest(c *gin.Context, info *relaycommon.RelayInfo, request *dto.GeneralOpenAIRequest) (any, error) {
+	if request == nil {
+		return nil, errors.New("request is nil")
+	}
+	switch info.RelayMode {
+	case constant.RelayModeEmbeddings:
+		baiduEmbeddingRequest := embeddingRequestOpenAI2Moka(*request)
+		return baiduEmbeddingRequest, nil
+	default:
+		return nil, errors.New("not implemented")
+	}
+}
+
+func (a *Adaptor) ConvertRerankRequest(c *gin.Context, relayMode int, request dto.RerankRequest) (any, error) {
+	return nil, nil
+}
+
+func (a *Adaptor) DoRequest(c *gin.Context, info *relaycommon.RelayInfo, requestBody io.Reader) (any, error) {
+	return channel.DoApiRequest(a, c, info, requestBody)
+}
+
+func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, info *relaycommon.RelayInfo) (usage any, err *dto.OpenAIErrorWithStatusCode) {
+		
+	switch info.RelayMode {
+	case constant.RelayModeEmbeddings:
+		err, usage = mokaEmbeddingHandler(c, resp)
+	default:
+		// err, usage = mokaHandler(c, resp)
+		
+	}
+	return
+}
+
+func (a *Adaptor) GetModelList() []string {
+	return ModelList
+}
+
+func (a *Adaptor) GetChannelName() string {
+	return ChannelName
+}

relay/channel/mokaai/constants.go

@@ -0,0 +1,9 @@
+package mokaai
+
+var ModelList = []string{
+	"m3e-large",
+	"m3e-base",
+	"m3e-small",
+}
+
+var ChannelName = "mokaai"

relay/channel/mokaai/relay-mokaai.go

@@ -0,0 +1,83 @@
+package mokaai
+
+import (
+	"encoding/json"
+	"github.com/gin-gonic/gin"
+	"io"
+	"net/http"
+	"one-api/dto"
+	"one-api/service"
+)
+
+func embeddingRequestOpenAI2Moka(request dto.GeneralOpenAIRequest) *dto.EmbeddingRequest {
+	var input []string // Change input to []string
+
+	switch v := request.Input.(type) {
+	case string:
+		input = []string{v} // Convert string to []string
+	case []string:
+		input = v // Already a []string, no conversion needed
+	case []interface{}:
+		for _, part := range v {
+			if str, ok := part.(string); ok {
+				input = append(input, str) // Append each string to the slice
+			}
+		}
+	}
+	return &dto.EmbeddingRequest{
+		Input: input,
+		Model:  request.Model,
+	}
+}
+
+func embeddingResponseMoka2OpenAI(response *dto.EmbeddingResponse) *dto.OpenAIEmbeddingResponse {
+	openAIEmbeddingResponse := dto.OpenAIEmbeddingResponse{
+		Object: "list",
+		Data:   make([]dto.OpenAIEmbeddingResponseItem, 0, len(response.Data)),
+		Model:  "baidu-embedding",
+		Usage:  response.Usage,
+	}
+	for _, item := range response.Data {
+		openAIEmbeddingResponse.Data = append(openAIEmbeddingResponse.Data, dto.OpenAIEmbeddingResponseItem{
+			Object:    item.Object,
+			Index:     item.Index,
+			Embedding: item.Embedding,
+		})
+	}
+	return &openAIEmbeddingResponse
+}
+
+func mokaEmbeddingHandler(c *gin.Context, resp *http.Response) (*dto.OpenAIErrorWithStatusCode, *dto.Usage) {
+	var baiduResponse dto.EmbeddingResponse
+	responseBody, err := io.ReadAll(resp.Body)
+	if err != nil {
+		return service.OpenAIErrorWrapper(err, "read_response_body_failed", http.StatusInternalServerError), nil
+	}
+	err = resp.Body.Close()
+	if err != nil {
+		return service.OpenAIErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), nil
+	}
+	err = json.Unmarshal(responseBody, &baiduResponse)
+	if err != nil {
+		return service.OpenAIErrorWrapper(err, "unmarshal_response_body_failed", http.StatusInternalServerError), nil
+	}
+	// if baiduResponse.ErrorMsg != "" {
+	// 	return &dto.OpenAIErrorWithStatusCode{
+	// 		Error: dto.OpenAIError{
+	// 			Type:    "baidu_error",
+	// 			Param:   "",
+	// 		},
+	// 		StatusCode: resp.StatusCode,
+	// 	}, nil
+	// }
+	fullTextResponse := embeddingResponseMoka2OpenAI(&baiduResponse)
+	jsonResponse, err := json.Marshal(fullTextResponse)
+	if err != nil {
+		return service.OpenAIErrorWrapper(err, "marshal_response_body_failed", http.StatusInternalServerError), nil
+	}
+	c.Writer.Header().Set("Content-Type", "application/json")
+	c.Writer.WriteHeader(resp.StatusCode)
+	_, err = c.Writer.Write(jsonResponse)
+	return nil, &fullTextResponse.Usage
+}
+

relay/channel/ollama/adaptor.go

@@ -39,6 +39,7 @@ func (a *Adaptor) GetRequestURL(info *relaycommon.RelayInfo) (string, error) {
 
 func (a *Adaptor) SetupRequestHeader(c *gin.Context, req *http.Header, info *relaycommon.RelayInfo) error {
 	channel.SetupApiRequestHeader(info, c, req)
+	req.Set("Authorization", "Bearer "+info.ApiKey)
 	return nil
 }
 
@@ -46,18 +47,17 @@ func (a *Adaptor) ConvertRequest(c *gin.Context, info *relaycommon.RelayInfo, re
 	if request == nil {
 		return nil, errors.New("request is nil")
 	}
-	switch info.RelayMode {
-	case relayconstant.RelayModeEmbeddings:
-		return requestOpenAI2Embeddings(*request), nil
-	default:
-		return requestOpenAI2Ollama(*request), nil
-	}
+	return requestOpenAI2Ollama(*request)
 }
 
 func (a *Adaptor) ConvertRerankRequest(c *gin.Context, relayMode int, request dto.RerankRequest) (any, error) {
 	return nil, nil
 }
 
+func (a *Adaptor) ConvertEmbeddingRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.EmbeddingRequest) (any, error) {
+	return requestOpenAI2Embeddings(request), nil
+}
+
 func (a *Adaptor) DoRequest(c *gin.Context, info *relaycommon.RelayInfo, requestBody io.Reader) (any, error) {
 	return channel.DoApiRequest(a, c, info, requestBody)
 }

relay/channel/ollama/dto.go

@@ -3,18 +3,22 @@ package ollama
 import "one-api/dto"
 
 type OllamaRequest struct {
-	Model            string         `json:"model,omitempty"`
-	Messages         []dto.Message  `json:"messages,omitempty"`
-	Stream           bool           `json:"stream,omitempty"`
-	Temperature      *float64       `json:"temperature,omitempty"`
-	Seed             float64        `json:"seed,omitempty"`
-	Topp             float64        `json:"top_p,omitempty"`
-	TopK             int            `json:"top_k,omitempty"`
-	Stop             any            `json:"stop,omitempty"`
-	Tools            []dto.ToolCall `json:"tools,omitempty"`
-	ResponseFormat   any            `json:"response_format,omitempty"`
-	FrequencyPenalty float64        `json:"frequency_penalty,omitempty"`
-	PresencePenalty  float64        `json:"presence_penalty,omitempty"`
+	Model            string                `json:"model,omitempty"`
+	Messages         []dto.Message         `json:"messages,omitempty"`
+	Stream           bool                  `json:"stream,omitempty"`
+	Temperature      *float64              `json:"temperature,omitempty"`
+	Seed             float64               `json:"seed,omitempty"`
+	Topp             float64               `json:"top_p,omitempty"`
+	TopK             int                   `json:"top_k,omitempty"`
+	Stop             any                   `json:"stop,omitempty"`
+	MaxTokens        uint                  `json:"max_tokens,omitempty"`
+	Tools            []dto.ToolCallRequest `json:"tools,omitempty"`
+	ResponseFormat   any                   `json:"response_format,omitempty"`
+	FrequencyPenalty float64               `json:"frequency_penalty,omitempty"`
+	PresencePenalty  float64               `json:"presence_penalty,omitempty"`
+	Suffix           any                   `json:"suffix,omitempty"`
+	StreamOptions    *dto.StreamOptions    `json:"stream_options,omitempty"`
+	Prompt           any                   `json:"prompt,omitempty"`
 }
 
 type Options struct {
@@ -35,7 +39,7 @@ type OllamaEmbeddingRequest struct {
 }
 
 type OllamaEmbeddingResponse struct {
-	Error     string    `json:"error,omitempty"`
-	Model     string    `json:"model"`
+	Error     string      `json:"error,omitempty"`
+	Model     string      `json:"model"`
 	Embedding [][]float64 `json:"embeddings,omitempty"`
 }

relay/channel/ollama/relay-ollama.go

@@ -9,14 +9,36 @@ import (
 	"net/http"
 	"one-api/dto"
 	"one-api/service"
+	"strings"
 )
 
-func requestOpenAI2Ollama(request dto.GeneralOpenAIRequest) *OllamaRequest {
+func requestOpenAI2Ollama(request dto.GeneralOpenAIRequest) (*OllamaRequest, error) {
 	messages := make([]dto.Message, 0, len(request.Messages))
 	for _, message := range request.Messages {
+		if !message.IsStringContent() {
+			mediaMessages := message.ParseContent()
+			for j, mediaMessage := range mediaMessages {
+				if mediaMessage.Type == dto.ContentTypeImageURL {
+					imageUrl := mediaMessage.ImageUrl.(dto.MessageImageUrl)
+					// check if not base64
+					if strings.HasPrefix(imageUrl.Url, "http") {
+						fileData, err := service.GetFileBase64FromUrl(imageUrl.Url)
+						if err != nil {
+							return nil, err
+						}
+						imageUrl.Url = fmt.Sprintf("data:%s;base64,%s", fileData.MimeType, fileData.Base64Data)
+					}
+					mediaMessage.ImageUrl = imageUrl
+					mediaMessages[j] = mediaMessage
+				}
+			}
+			message.SetMediaContent(mediaMessages)
+		}
 		messages = append(messages, dto.Message{
-			Role:    message.Role,
-			Content: message.Content,
+			Role:       message.Role,
+			Content:    message.Content,
+			ToolCalls:  message.ToolCalls,
+			ToolCallId: message.ToolCallId,
 		})
 	}
 	str, ok := request.Stop.(string)
@@ -36,13 +58,17 @@ func requestOpenAI2Ollama(request dto.GeneralOpenAIRequest) *OllamaRequest {
 		TopK:             request.TopK,
 		Stop:             Stop,
 		Tools:            request.Tools,
+		MaxTokens:        request.MaxTokens,
 		ResponseFormat:   request.ResponseFormat,
 		FrequencyPenalty: request.FrequencyPenalty,
 		PresencePenalty:  request.PresencePenalty,
-	}
+		Prompt:           request.Prompt,
+		StreamOptions:    request.StreamOptions,
+		Suffix:           request.Suffix,
+	}, nil
 }
 
-func requestOpenAI2Embeddings(request dto.GeneralOpenAIRequest) *OllamaEmbeddingRequest {
+func requestOpenAI2Embeddings(request dto.EmbeddingRequest) *OllamaEmbeddingRequest {
 	return &OllamaEmbeddingRequest{
 		Model: request.Model,
 		Input: request.ParseInput(),
@@ -123,9 +149,9 @@ func ollamaEmbeddingHandler(c *gin.Context, resp *http.Response, promptTokens in
 }
 
 func flattenEmbeddings(embeddings [][]float64) []float64 {
-flattened := []float64{}
-for _, row := range embeddings {
-	flattened = append(flattened, row...)
+	flattened := []float64{}
+	for _, row := range embeddings {
+		flattened = append(flattened, row...)
+	}
+	return flattened
 }
-return flattened
-}

relay/channel/openai/adaptor.go

@@ -10,9 +10,11 @@ import (
 	"mime/multipart"
 	"net/http"
 	"one-api/common"
+	constant2 "one-api/constant"
 	"one-api/dto"
 	"one-api/relay/channel"
 	"one-api/relay/channel/ai360"
+	"one-api/relay/channel/jina"
 	"one-api/relay/channel/lingyiwanwu"
 	"one-api/relay/channel/minimax"
 	"one-api/relay/channel/moonshot"
@@ -44,16 +46,20 @@ func (a *Adaptor) GetRequestURL(info *relaycommon.RelayInfo) (string, error) {
 	}
 	switch info.ChannelType {
 	case common.ChannelTypeAzure:
+		apiVersion := info.ApiVersion
+		if apiVersion == "" {
+			apiVersion = constant2.AzureDefaultAPIVersion
+		}
 		// https://learn.microsoft.com/en-us/azure/cognitive-services/openai/chatgpt-quickstart?pivots=rest-api&tabs=command-line#rest-api
 		requestURL := strings.Split(info.RequestURLPath, "?")[0]
-		requestURL = fmt.Sprintf("%s?api-version=%s", requestURL, info.ApiVersion)
+		requestURL = fmt.Sprintf("%s?api-version=%s", requestURL, apiVersion)
 		task := strings.TrimPrefix(requestURL, "/v1/")
 		model_ := info.UpstreamModelName
 		model_ = strings.Replace(model_, ".", "", -1)
 		// https://github.com/songquanpeng/one-api/issues/67
 		requestURL = fmt.Sprintf("/openai/deployments/%s/%s", model_, task)
 		if info.RelayMode == constant.RelayModeRealtime {
-			requestURL = fmt.Sprintf("/openai/realtime?deployment=%s&api-version=%s", model_, info.ApiVersion)
+			requestURL = fmt.Sprintf("/openai/realtime?deployment=%s&api-version=%s", model_, apiVersion)
 		}
 		return relaycommon.GetFullRequestURL(info.BaseUrl, requestURL, info.ChannelType), nil
 	case common.ChannelTypeMiniMax:
@@ -109,13 +115,28 @@ func (a *Adaptor) ConvertRequest(c *gin.Context, info *relaycommon.RelayInfo, re
 	if info.ChannelType != common.ChannelTypeOpenAI && info.ChannelType != common.ChannelTypeAzure {
 		request.StreamOptions = nil
 	}
-	if strings.HasPrefix(request.Model, "o1") {
+	if strings.HasPrefix(request.Model, "o1") || strings.HasPrefix(request.Model, "o3") {
 		if request.MaxCompletionTokens == 0 && request.MaxTokens != 0 {
 			request.MaxCompletionTokens = request.MaxTokens
 			request.MaxTokens = 0
 		}
+		if strings.HasPrefix(request.Model, "o3") || strings.HasPrefix(request.Model, "o1") {
+			request.Temperature = nil
+		}
+		if strings.HasSuffix(request.Model, "-high") {
+			request.ReasoningEffort = "high"
+			request.Model = strings.TrimSuffix(request.Model, "-high")
+		} else if strings.HasSuffix(request.Model, "-low") {
+			request.ReasoningEffort = "low"
+			request.Model = strings.TrimSuffix(request.Model, "-low")
+		} else if strings.HasSuffix(request.Model, "-medium") {
+			request.ReasoningEffort = "medium"
+			request.Model = strings.TrimSuffix(request.Model, "-medium")
+		}
+		info.ReasoningEffort = request.ReasoningEffort
+		info.UpstreamModelName = request.Model
 	}
-	if request.Model == "o1" || request.Model == "o1-2024-12-17" {
+	if request.Model == "o1" || request.Model == "o1-2024-12-17" || strings.HasPrefix(request.Model, "o3") {
 		//修改第一个Message的内容，将system改为developer
 		if len(request.Messages) > 0 && request.Messages[0].Role == "system" {
 			request.Messages[0].Role = "developer"
@@ -126,7 +147,11 @@ func (a *Adaptor) ConvertRequest(c *gin.Context, info *relaycommon.RelayInfo, re
 }
 
 func (a *Adaptor) ConvertRerankRequest(c *gin.Context, relayMode int, request dto.RerankRequest) (any, error) {
-	return nil, errors.New("not implemented")
+	return request, nil
+}
+
+func (a *Adaptor) ConvertEmbeddingRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.EmbeddingRequest) (any, error) {
+	return request, nil
 }
 
 func (a *Adaptor) ConvertAudioRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.AudioRequest) (io.Reader, error) {
@@ -204,6 +229,8 @@ func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, info *relaycom
 		err, usage = OpenaiSTTHandler(c, resp, info, a.ResponseFormat)
 	case constant.RelayModeImagesGenerations:
 		err, usage = OpenaiTTSHandler(c, resp, info)
+	case constant.RelayModeRerank:
+		err, usage = jina.JinaRerankHandler(c, resp)
 	default:
 		if info.IsStream {
 			err, usage = OaiStreamHandler(c, resp, info)

relay/channel/openai/constant.go

@@ -11,8 +11,13 @@ var ModelList = []string{
 	"chatgpt-4o-latest",
 	"gpt-4o", "gpt-4o-2024-05-13", "gpt-4o-2024-08-06", "gpt-4o-2024-11-20",
 	"gpt-4o-mini", "gpt-4o-mini-2024-07-18",
+	"gpt-4.5-preview", "gpt-4.5-preview-2025-02-27",
 	"o1-preview", "o1-preview-2024-09-12",
 	"o1-mini", "o1-mini-2024-09-12",
+	"o3-mini", "o3-mini-2025-01-31",
+	"o3-mini-high", "o3-mini-2025-01-31-high",
+	"o3-mini-low", "o3-mini-2025-01-31-low",
+	"o3-mini-medium", "o3-mini-2025-01-31-medium",
 	"o1", "o1-2024-12-17",
 	"gpt-4o-audio-preview", "gpt-4o-audio-preview-2024-10-01",
 	"gpt-4o-realtime-preview", "gpt-4o-realtime-preview-2024-10-01", "gpt-4o-realtime-preview-2024-12-17",

relay/channel/openai/relay-openai.go

@@ -1,11 +1,9 @@
 package openai
 
 import (
-	"bufio"
 	"bytes"
 	"encoding/json"
 	"fmt"
-	"github.com/pkg/errors"
 	"io"
 	"math"
 	"mime/multipart"
@@ -15,30 +13,94 @@ import (
 	"one-api/dto"
 	relaycommon "one-api/relay/common"
 	relayconstant "one-api/relay/constant"
+	"one-api/relay/helper"
 	"one-api/service"
 	"os"
 	"strings"
-	"sync"
-	"time"
 
 	"github.com/bytedance/gopkg/util/gopool"
 	"github.com/gin-gonic/gin"
 	"github.com/gorilla/websocket"
+	"github.com/pkg/errors"
 )
 
-func sendStreamData(c *gin.Context, data string, forceFormat bool) error {
+func sendStreamData(c *gin.Context, info *relaycommon.RelayInfo, data string, forceFormat bool, thinkToContent bool) error {
 	if data == "" {
 		return nil
 	}
 
-	if forceFormat {
-		var lastStreamResponse dto.ChatCompletionsStreamResponse
-		if err := json.Unmarshal(common.StringToByteSlice(data), &lastStreamResponse); err != nil {
-			return err
-		}
-		return service.ObjectData(c, lastStreamResponse)
+	if !forceFormat && !thinkToContent {
+		return helper.StringData(c, data)
 	}
-	return service.StringData(c, data)
+
+	var lastStreamResponse dto.ChatCompletionsStreamResponse
+	if err := json.Unmarshal(common.StringToByteSlice(data), &lastStreamResponse); err != nil {
+		return err
+	}
+
+	if !thinkToContent {
+		return helper.ObjectData(c, lastStreamResponse)
+	}
+
+	hasThinkingContent := false
+	hasContent := false
+	var thinkingContent strings.Builder
+	for _, choice := range lastStreamResponse.Choices {
+		if len(choice.Delta.GetReasoningContent()) > 0 {
+			hasThinkingContent = true
+			thinkingContent.WriteString(choice.Delta.GetReasoningContent())
+		}
+		if len(choice.Delta.GetContentString()) > 0 {
+			hasContent = true
+		}
+	}
+
+	// Handle think to content conversion
+	if info.ThinkingContentInfo.IsFirstThinkingContent {
+		if hasThinkingContent {
+			response := lastStreamResponse.Copy()
+			for i := range response.Choices {
+				// send `think` tag with thinking content
+				response.Choices[i].Delta.SetContentString("<think>\n" + thinkingContent.String())
+				response.Choices[i].Delta.ReasoningContent = nil
+				response.Choices[i].Delta.Reasoning = nil
+			}
+			info.ThinkingContentInfo.IsFirstThinkingContent = false
+			return helper.ObjectData(c, response)
+		}
+	}
+
+	if lastStreamResponse.Choices == nil || len(lastStreamResponse.Choices) == 0 {
+		return helper.ObjectData(c, lastStreamResponse)
+	}
+
+	// Process each choice
+	for i, choice := range lastStreamResponse.Choices {
+		// Handle transition from thinking to content
+		if hasContent && !info.ThinkingContentInfo.SendLastThinkingContent {
+			response := lastStreamResponse.Copy()
+			for j := range response.Choices {
+				response.Choices[j].Delta.SetContentString("\n</think>\n")
+				response.Choices[j].Delta.ReasoningContent = nil
+				response.Choices[j].Delta.Reasoning = nil
+			}
+			info.ThinkingContentInfo.SendLastThinkingContent = true
+			helper.ObjectData(c, response)
+		}
+
+		// Convert reasoning content to regular content
+		if len(choice.Delta.GetReasoningContent()) > 0 {
+			lastStreamResponse.Choices[i].Delta.SetContentString(choice.Delta.GetReasoningContent())
+			lastStreamResponse.Choices[i].Delta.ReasoningContent = nil
+			lastStreamResponse.Choices[i].Delta.Reasoning = nil
+		} else if !hasThinkingContent && !hasContent {
+			// flush thinking content
+			lastStreamResponse.Choices[i].Delta.ReasoningContent = nil
+			lastStreamResponse.Choices[i].Delta.Reasoning = nil
+		}
+	}
+
+	return helper.ObjectData(c, lastStreamResponse)
 }
 
 func OaiStreamHandler(c *gin.Context, resp *http.Response, info *relaycommon.RelayInfo) (*dto.OpenAIErrorWithStatusCode, *dto.Usage) {
@@ -57,67 +119,33 @@ func OaiStreamHandler(c *gin.Context, resp *http.Response, info *relaycommon.Rel
 	var usage = &dto.Usage{}
 	var streamItems []string // store stream items
 	var forceFormat bool
+	var thinkToContent bool
 
-	if info.ChannelType == common.ChannelTypeCustom {
-		if forceFmt, ok := info.ChannelSetting["force_format"].(bool); ok {
-			forceFormat = forceFmt
-		}
+	if forceFmt, ok := info.ChannelSetting[constant.ForceFormat].(bool); ok {
+		forceFormat = forceFmt
+	}
+
+	if think2Content, ok := info.ChannelSetting[constant.ChannelSettingThinkingToContent].(bool); ok {
+		thinkToContent = think2Content
 	}
 
 	toolCount := 0
-	scanner := bufio.NewScanner(resp.Body)
-	scanner.Split(bufio.ScanLines)
 
-	service.SetEventStreamHeaders(c)
-	streamingTimeout := time.Duration(constant.StreamingTimeout) * time.Second
-	if strings.HasPrefix(info.UpstreamModelName, "o1") || strings.HasPrefix(info.UpstreamModelName, "o3") {
-		// twice timeout for o1 model
-		streamingTimeout *= 2
-	}
-	ticker := time.NewTicker(streamingTimeout)
-	defer ticker.Stop()
-
-	stopChan := make(chan bool)
-	defer close(stopChan)
 	var (
 		lastStreamData string
-		mu             sync.Mutex
 	)
-	gopool.Go(func() {
-		for scanner.Scan() {
-			info.SetFirstResponseTime()
-			ticker.Reset(time.Duration(constant.StreamingTimeout) * time.Second)
-			data := scanner.Text()
-			if len(data) < 6 { // ignore blank line or wrong format
-				continue
-			}
-			if data[:6] != "data: " && data[:6] != "[DONE]" {
-				continue
-			}
-			mu.Lock()
-			data = data[6:]
-			if !strings.HasPrefix(data, "[DONE]") {
-				if lastStreamData != "" {
-					err := sendStreamData(c, lastStreamData, forceFormat)
-					if err != nil {
-						common.LogError(c, "streaming error: "+err.Error())
-					}
-				}
-				lastStreamData = data
-				streamItems = append(streamItems, data)
-			}
-			mu.Unlock()
-		}
-		common.SafeSendBool(stopChan, true)
-	})
 
-	select {
-	case <-ticker.C:
-		// 超时处理逻辑
-		common.LogError(c, "streaming timeout")
-	case <-stopChan:
-		// 正常结束
-	}
+	helper.StreamScannerHandler(c, resp, info, func(data string) bool {
+		if lastStreamData != "" {
+			err := sendStreamData(c, info, lastStreamData, forceFormat, thinkToContent)
+			if err != nil {
+				common.LogError(c, "streaming error: "+err.Error())
+			}
+		}
+		lastStreamData = data
+		streamItems = append(streamItems, data)
+		return true
+	})
 
 	shouldSendLastResp := true
 	var lastStreamResponse dto.ChatCompletionsStreamResponse
@@ -141,7 +169,7 @@ func OaiStreamHandler(c *gin.Context, resp *http.Response, info *relaycommon.Rel
 		}
 	}
 	if shouldSendLastResp {
-		sendStreamData(c, lastStreamData, forceFormat)
+		sendStreamData(c, info, lastStreamData, forceFormat, thinkToContent)
 	}
 
 	// 计算token
@@ -162,6 +190,10 @@ func OaiStreamHandler(c *gin.Context, resp *http.Response, info *relaycommon.Rel
 					//}
 					for _, choice := range streamResponse.Choices {
 						responseTextBuilder.WriteString(choice.Delta.GetContentString())
+
+						// handle both reasoning_content and reasoning
+						responseTextBuilder.WriteString(choice.Delta.GetReasoningContent())
+
 						if choice.Delta.ToolCalls != nil {
 							if len(choice.Delta.ToolCalls) > toolCount {
 								toolCount = len(choice.Delta.ToolCalls)
@@ -182,6 +214,7 @@ func OaiStreamHandler(c *gin.Context, resp *http.Response, info *relaycommon.Rel
 				//}
 				for _, choice := range streamResponse.Choices {
 					responseTextBuilder.WriteString(choice.Delta.GetContentString())
+					responseTextBuilder.WriteString(choice.Delta.GetReasoningContent()) // This will handle both reasoning_content and reasoning
 					if choice.Delta.ToolCalls != nil {
 						if len(choice.Delta.ToolCalls) > toolCount {
 							toolCount = len(choice.Delta.ToolCalls)
@@ -221,17 +254,23 @@ func OaiStreamHandler(c *gin.Context, resp *http.Response, info *relaycommon.Rel
 	if !containStreamUsage {
 		usage, _ = service.ResponseText2Usage(responseTextBuilder.String(), info.UpstreamModelName, info.PromptTokens)
 		usage.CompletionTokens += toolCount * 7
+	} else {
+		if info.ChannelType == common.ChannelTypeDeepSeek {
+			if usage.PromptCacheHitTokens != 0 {
+				usage.PromptTokensDetails.CachedTokens = usage.PromptCacheHitTokens
+			}
+		}
 	}
 
 	if info.ShouldIncludeUsage && !containStreamUsage {
-		response := service.GenerateFinalUsageResponse(responseId, createAt, model, *usage)
+		response := helper.GenerateFinalUsageResponse(responseId, createAt, model, *usage)
 		response.SetSystemFingerprint(systemFingerprint)
-		service.ObjectData(c, response)
+		helper.ObjectData(c, response)
 	}
 
-	service.Done(c)
+	helper.Done(c)
 
-	resp.Body.Close()
+	//resp.Body.Close()
 	return nil, usage
 }
 
@@ -273,7 +312,7 @@ func OpenaiHandler(c *gin.Context, resp *http.Response, promptTokens int, model
 	if simpleResponse.Usage.TotalTokens == 0 || (simpleResponse.Usage.PromptTokens == 0 && simpleResponse.Usage.CompletionTokens == 0) {
 		completionTokens := 0
 		for _, choice := range simpleResponse.Choices {
-			ctkm, _ := service.CountTextToken(string(choice.Message.Content), model)
+			ctkm, _ := service.CountTextToken(choice.Message.StringContent()+choice.Message.ReasoningContent+choice.Message.Reasoning, model)
 			completionTokens += ctkm
 		}
 		simpleResponse.Usage = dto.Usage{
@@ -462,7 +501,7 @@ func OpenaiRealtimeHandler(c *gin.Context, info *relaycommon.RelayInfo) (*dto.Op
 				localUsage.InputTokenDetails.TextTokens += textToken
 				localUsage.InputTokenDetails.AudioTokens += audioToken
 
-				err = service.WssString(c, targetConn, string(message))
+				err = helper.WssString(c, targetConn, string(message))
 				if err != nil {
 					errChan <- fmt.Errorf("error writing to target: %v", err)
 					return
@@ -568,7 +607,7 @@ func OpenaiRealtimeHandler(c *gin.Context, info *relaycommon.RelayInfo) (*dto.Op
 					localUsage.OutputTokenDetails.AudioTokens += audioToken
 				}
 
-				err = service.WssString(c, clientConn, string(message))
+				err = helper.WssString(c, clientConn, string(message))
 				if err != nil {
 					errChan <- fmt.Errorf("error writing to client: %v", err)
 					return

relay/channel/openrouter/adaptor.go

@@ -0,0 +1,74 @@
+package openrouter
+
+import (
+	"errors"
+	"fmt"
+	"github.com/gin-gonic/gin"
+	"io"
+	"net/http"
+	"one-api/dto"
+	"one-api/relay/channel"
+	"one-api/relay/channel/openai"
+	relaycommon "one-api/relay/common"
+)
+
+type Adaptor struct {
+}
+
+func (a *Adaptor) ConvertAudioRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.AudioRequest) (io.Reader, error) {
+	//TODO implement me
+	return nil, errors.New("not implemented")
+}
+
+func (a *Adaptor) ConvertImageRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.ImageRequest) (any, error) {
+	//TODO implement me
+	return nil, errors.New("not implemented")
+}
+
+func (a *Adaptor) Init(info *relaycommon.RelayInfo) {
+}
+
+func (a *Adaptor) GetRequestURL(info *relaycommon.RelayInfo) (string, error) {
+	return fmt.Sprintf("%s/v1/chat/completions", info.BaseUrl), nil
+}
+
+func (a *Adaptor) SetupRequestHeader(c *gin.Context, req *http.Header, info *relaycommon.RelayInfo) error {
+	channel.SetupApiRequestHeader(info, c, req)
+	req.Set("Authorization", fmt.Sprintf("Bearer %s", info.ApiKey))
+	req.Set("HTTP-Referer", "https://github.com/Calcium-Ion/new-api")
+	req.Set("X-Title", "New API")
+	return nil
+}
+
+func (a *Adaptor) ConvertRequest(c *gin.Context, info *relaycommon.RelayInfo, request *dto.GeneralOpenAIRequest) (any, error) {
+	return request, nil
+}
+
+func (a *Adaptor) DoRequest(c *gin.Context, info *relaycommon.RelayInfo, requestBody io.Reader) (any, error) {
+	return channel.DoApiRequest(a, c, info, requestBody)
+}
+
+func (a *Adaptor) ConvertRerankRequest(c *gin.Context, relayMode int, request dto.RerankRequest) (any, error) {
+	return nil, errors.New("not implemented")
+}
+
+func (a *Adaptor) ConvertEmbeddingRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.EmbeddingRequest) (any, error) {
+	return nil, errors.New("not implemented")
+}
+
+func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, info *relaycommon.RelayInfo) (usage any, err *dto.OpenAIErrorWithStatusCode) {
+	if info.IsStream {
+		err, usage = openai.OaiStreamHandler(c, resp, info)
+	} else {
+		err, usage = openai.OpenaiHandler(c, resp, info.PromptTokens, info.UpstreamModelName)
+	}
+	return
+}
+
+func (a *Adaptor) GetModelList() []string {
+	return ModelList
+}
+
+func (a *Adaptor) GetChannelName() string {
+	return ChannelName
+}

relay/channel/openrouter/constant.go

@@ -0,0 +1,5 @@
+package openrouter
+
+var ModelList = []string{}
+
+var ChannelName = "openrouter"

relay/channel/palm/adaptor.go

@@ -49,6 +49,12 @@ func (a *Adaptor) ConvertRerankRequest(c *gin.Context, relayMode int, request dt
 	return nil, nil
 }
 
+func (a *Adaptor) ConvertEmbeddingRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.EmbeddingRequest) (any, error) {
+	//TODO implement me
+	return nil, errors.New("not implemented")
+}
+
+
 func (a *Adaptor) DoRequest(c *gin.Context, info *relaycommon.RelayInfo, requestBody io.Reader) (any, error) {
 	return channel.DoApiRequest(a, c, info, requestBody)
 }

relay/channel/palm/relay-palm.go

@@ -9,6 +9,7 @@ import (
 	"one-api/common"
 	"one-api/constant"
 	"one-api/dto"
+	"one-api/relay/helper"
 	"one-api/service"
 )
 
@@ -112,7 +113,7 @@ func palmStreamHandler(c *gin.Context, resp *http.Response) (*dto.OpenAIErrorWit
 		dataChan <- string(jsonResponse)
 		stopChan <- true
 	}()
-	service.SetEventStreamHeaders(c)
+	helper.SetEventStreamHeaders(c)
 	c.Stream(func(w io.Writer) bool {
 		select {
 		case data := <-dataChan:

relay/channel/perplexity/adaptor.go

@@ -52,6 +52,12 @@ func (a *Adaptor) ConvertRerankRequest(c *gin.Context, relayMode int, request dt
 	return nil, nil
 }
 
+func (a *Adaptor) ConvertEmbeddingRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.EmbeddingRequest) (any, error) {
+	//TODO implement me
+	return nil, errors.New("not implemented")
+}
+
+
 func (a *Adaptor) DoRequest(c *gin.Context, info *relaycommon.RelayInfo, requestBody io.Reader) (any, error) {
 	return channel.DoApiRequest(a, c, info, requestBody)
 }

relay/channel/siliconflow/adaptor.go

@@ -36,6 +36,8 @@ func (a *Adaptor) GetRequestURL(info *relaycommon.RelayInfo) (string, error) {
 		return fmt.Sprintf("%s/v1/embeddings", info.BaseUrl), nil
 	} else if info.RelayMode == constant.RelayModeChatCompletions {
 		return fmt.Sprintf("%s/v1/chat/completions", info.BaseUrl), nil
+	} else if info.RelayMode == constant.RelayModeCompletions {
+		return fmt.Sprintf("%s/v1/completions", info.BaseUrl), nil
 	}
 	return "", errors.New("invalid relay mode")
 }
@@ -58,6 +60,10 @@ func (a *Adaptor) ConvertRerankRequest(c *gin.Context, relayMode int, request dt
 	return request, nil
 }
 
+func (a *Adaptor) ConvertEmbeddingRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.EmbeddingRequest) (any, error) {
+	return request, nil
+}
+
 func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, info *relaycommon.RelayInfo) (usage any, err *dto.OpenAIErrorWithStatusCode) {
 	switch info.RelayMode {
 	case constant.RelayModeRerank:
@@ -68,6 +74,12 @@ func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, info *relaycom
 		} else {
 			err, usage = openai.OpenaiHandler(c, resp, info.PromptTokens, info.UpstreamModelName)
 		}
+	case constant.RelayModeCompletions:
+		if info.IsStream {
+			err, usage = openai.OaiStreamHandler(c, resp, info)
+		} else {
+			err, usage = openai.OpenaiHandler(c, resp, info.PromptTokens, info.UpstreamModelName)
+		}
 	case constant.RelayModeEmbeddings:
 		err, usage = openai.OpenaiHandler(c, resp, info.PromptTokens, info.UpstreamModelName)
 	}

relay/channel/siliconflow/constant.go

@@ -40,7 +40,7 @@ var ModelList = []string{
 	"Pro/meta-llama/Meta-Llama-3-8B-Instruct",
 	"Pro/mistralai/Mistral-7B-Instruct-v0.2",
 	"black-forest-labs/FLUX.1-schnell",
-	"iic/SenseVoiceSmall",
+	"FunAudioLLM/SenseVoiceSmall",
 	"netease-youdao/bce-embedding-base_v1",
 	"BAAI/bge-m3",
 	"internlm/internlm2_5-20b-chat",