Merge pull request #820 from Calcium-Ion/cache

feat: Implement cache token ratio for more precise token pricing

.env.example

@@ -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

README.en.md

@@ -68,7 +68,7 @@
 
 ## 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)
@@ -162,7 +162,7 @@ docker run --rm -v /var/run/docker.sock:/var/run/docker.sock containrrr/watchtow
 
 ## 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

@@ -77,7 +77,7 @@
 
 ## 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)
@@ -94,7 +94,6 @@
 - `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`。
 - `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`。
@@ -103,6 +102,10 @@
 - `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]
@@ -174,7 +177,7 @@ docker run --rm -v /var/run/docker.sock:/var/run/docker.sock containrrr/watchtow
 
 ## 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`
@@ -217,8 +220,8 @@ docker run --rm -v /var/run/docker.sock:/var/run/docker.sock containrrr/watchtow
 - [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
 

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)
+}

constant/env.go

@@ -1,10 +1,7 @@
 package constant
 
 import (
-	"fmt"
 	"one-api/common"
-	"os"
-	"strings"
 )
 
 var StreamingTimeout = common.GetEnvOrDefault("STREAMING_TIMEOUT", 60)
@@ -23,9 +20,9 @@ var UpdateTask = common.GetEnvOrDefaultBool("UPDATE_TASK", true)
 
 var AzureDefaultAPIVersion = common.GetEnvOrDefaultString("AZURE_DEFAULT_API_VERSION", "2024-12-01-preview")
 
-var GeminiModelMap = map[string]string{
-	"gemini-1.0-pro": "v1",
-}
+//var GeminiModelMap = map[string]string{
+//	"gemini-1.0-pro": "v1",
+//}
 
 var GeminiVisionMaxImageNum = common.GetEnvOrDefault("GEMINI_VISION_MAX_IMAGE_NUM", 16)
 
@@ -33,18 +30,18 @@ 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))
+	//	}
+	//}
 }
 
 // GenerateDefaultToken 是否生成初始令牌，默认关闭。

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"
@@ -48,7 +49,7 @@ func testChannel(channel *model.Channel, testModel string) (err error, openAIErr
 	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
-		testModel == "text-embedding-v1" ||
+		strings.Contains(testModel, "embed") ||
 		channel.Type == common.ChannelTypeMokaAI { // 其他 embedding 模型
 		requestPath = "/v1/embeddings" // 修改请求路径
 	}
@@ -72,17 +73,11 @@ func testChannel(channel *model.Channel, testModel string) (err error, openAIErr
 		}
 	}
 
-	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]
-		}
+	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")
@@ -91,7 +86,14 @@ func testChannel(channel *model.Channel, testModel string) (err error, openAIErr
 
 	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 {
@@ -99,12 +101,11 @@ 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 , meta %v ", channel.Id, testModel, meta))
+	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
 	}
@@ -114,7 +115,7 @@ 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
 	}
@@ -126,7 +127,7 @@ 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
 	}
@@ -139,26 +140,27 @@ 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, 0, 0.0, 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
 }

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"
@@ -66,7 +67,8 @@ 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":        setting.DemoSiteEnabled,
+			"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) {
@@ -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"
 )
@@ -41,15 +42,6 @@ func relayHandler(c *gin.Context, relayMode int) *dto.OpenAIErrorWithStatusCode
 	return err
 }
 
-func wsHandler(c *gin.Context, ws *websocket.Conn, relayMode int) *dto.OpenAIErrorWithStatusCode {
-	var err *dto.OpenAIErrorWithStatusCode
-	switch relayMode {
-	default:
-		err = relay.TextHelper(c)
-	}
-	return err
-}
-
 func Relay(c *gin.Context) {
 	relayMode := constant.Path2RelayMode(c.Request.URL.Path)
 	requestId := c.GetString(common.RequestIdKey)
@@ -110,7 +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
 	}
 
@@ -152,7 +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)
 	}
 }
 

dto/openai_request.go

@@ -18,50 +18,52 @@ type FormatJsonSchema struct {
 }
 
 type GeneralOpenAIRequest struct {
-	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               []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"`
-	ExtraBody           any             `json:"extra_body,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 {
@@ -97,6 +99,7 @@ type Message struct {
 	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
@@ -137,11 +140,11 @@ func (m *Message) SetPrefix(prefix bool) {
 	m.Prefix = &prefix
 }
 
-func (m *Message) ParseToolCalls() []ToolCall {
+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
 	}

dto/openai_response.go

@@ -62,10 +62,11 @@ type ChatCompletionsStreamResponseChoice struct {
 }
 
 type ChatCompletionsStreamResponseChoiceDelta struct {
-	Content          *string    `json:"content,omitempty"`
-	ReasoningContent *string    `json:"reasoning_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) {
@@ -80,34 +81,38 @@ func (c *ChatCompletionsStreamResponseChoiceDelta) GetContentString() string {
 }
 
 func (c *ChatCompletionsStreamResponseChoiceDelta) GetReasoningContent() string {
-	if c.ReasoningContent == nil {
+	if c.ReasoningContent == nil && c.Reasoning == nil {
 		return ""
 	}
-	return *c.ReasoningContent
+	if c.ReasoningContent != nil {
+		return *c.ReasoningContent
+	}
+	return *c.Reasoning
 }
 
 func (c *ChatCompletionsStreamResponseChoiceDelta) SetReasoningContent(s string) {
 	c.ReasoningContent = &s
+	c.Reasoning = &s
 }
 
-type ToolCall struct {
+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 FunctionCall `json:"function"`
+	Index    *int             `json:"index,omitempty"`
+	ID       string           `json:"id,omitempty"`
+	Type     any              `json:"type"`
+	Function FunctionResponse `json:"function"`
 }
 
-func (c *ToolCall) SetIndex(i int) {
+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 {

middleware/model-rate-limit.go

@@ -51,7 +51,7 @@ func checkRedisRateLimit(ctx context.Context, rdb *redis.Client, key string, max
 	// 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, common.RateLimitKeyExpirationDuration)
+		rdb.Expire(ctx, key, time.Duration(setting.ModelRequestRateLimitDurationMinutes)*time.Minute)
 		return false, nil
 	}
 
@@ -68,7 +68,7 @@ func recordRedisRequest(ctx context.Context, rdb *redis.Client, key string, maxC
 	now := time.Now().Format(timeFormat)
 	rdb.LPush(ctx, key, now)
 	rdb.LTrim(ctx, key, 0, int64(maxCount-1))
-	rdb.Expire(ctx, key, common.RateLimitKeyExpirationDuration)
+	rdb.Expire(ctx, key, time.Duration(setting.ModelRequestRateLimitDurationMinutes)*time.Minute)
 }
 
 // Redis限流处理器
@@ -118,7 +118,7 @@ func redisRateLimitHandler(duration int64, totalMaxCount, successMaxCount int) g
 
 // 内存限流处理器
 func memoryRateLimitHandler(duration int64, totalMaxCount, successMaxCount int) gin.HandlerFunc {
-	inMemoryRateLimiter.Init(common.RateLimitKeyExpirationDuration)
+	inMemoryRateLimiter.Init(time.Duration(setting.ModelRequestRateLimitDurationMinutes) * time.Minute)
 
 	return func(c *gin.Context) {
 		userId := strconv.Itoa(c.GetInt("id"))
@@ -153,20 +153,23 @@ func memoryRateLimitHandler(duration int64, totalMaxCount, successMaxCount int)
 
 // ModelRequestRateLimit 模型请求限流中间件
 func ModelRequestRateLimit() func(c *gin.Context) {
-	// 如果未启用限流，直接放行
-	if !setting.ModelRequestRateLimitEnabled {
-		return defNext
-	}
+	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
+		// 计算限流参数
+		duration := int64(setting.ModelRequestRateLimitDurationMinutes * 60)
+		totalMaxCount := setting.ModelRequestRateLimitCount
+		successMaxCount := setting.ModelRequestRateLimitSuccessCount
 
-	// 根据存储类型选择限流处理器
-	if common.RedisEnabled {
-		return redisRateLimitHandler(duration, totalMaxCount, successMaxCount)
-	} else {
-		return memoryRateLimitHandler(duration, totalMaxCount, successMaxCount)
+		// 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/channel.go

@@ -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 {

model/log.go

@@ -2,12 +2,13 @@ package model
 
 import (
 	"fmt"
-	"github.com/gin-gonic/gin"
 	"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"`

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)
@@ -84,15 +88,16 @@ func InitOptionMap() {
 	common.OptionMap["QuotaForInviter"] = strconv.Itoa(common.QuotaForInviter)
 	common.OptionMap["QuotaForInvitee"] = strconv.Itoa(common.QuotaForInvitee)
 	common.OptionMap["QuotaRemindThreshold"] = strconv.Itoa(common.QuotaRemindThreshold)
-	common.OptionMap["ShouldPreConsumedQuota"] = strconv.Itoa(common.PreConsumedQuota)
+	common.OptionMap["PreConsumedQuota"] = strconv.Itoa(common.PreConsumedQuota)
 	common.OptionMap["ModelRequestRateLimitCount"] = strconv.Itoa(setting.ModelRequestRateLimitCount)
 	common.OptionMap["ModelRequestRateLimitDurationMinutes"] = strconv.Itoa(setting.ModelRequestRateLimitDurationMinutes)
 	common.OptionMap["ModelRequestRateLimitSuccessCount"] = strconv.Itoa(setting.ModelRequestRateLimitSuccessCount)
-	common.OptionMap["ModelRatio"] = common.ModelRatio2JSONString()
-	common.OptionMap["ModelPrice"] = common.ModelPrice2JSONString()
+	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
@@ -107,15 +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(setting.DemoSiteEnabled)
+	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"] = setting.AutomaticDisableKeywordsToString()
-	common.OptionMap["GeminiSafetySettings"] = setting.GeminiSafetySettingsJsonString()
+	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()
@@ -159,6 +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 {
@@ -228,14 +245,13 @@ func updateOptionMap(key string, value string) (err error) {
 		case "CheckSensitiveEnabled":
 			setting.CheckSensitiveEnabled = boolValue
 		case "DemoSiteEnabled":
-			setting.DemoSiteEnabled = boolValue
+			operation_setting.DemoSiteEnabled = boolValue
+		case "SelfUseModeEnabled":
+			operation_setting.SelfUseModeEnabled = boolValue
 		case "CheckSensitiveOnPromptEnabled":
 			setting.CheckSensitiveOnPromptEnabled = boolValue
 		case "ModelRequestRateLimitEnabled":
 			setting.ModelRequestRateLimitEnabled = boolValue
-
-		//case "CheckSensitiveOnCompletionEnabled":
-		//	constant.CheckSensitiveOnCompletionEnabled = boolValue
 		case "StopOnSensitiveEnabled":
 			setting.StopOnSensitiveEnabled = boolValue
 		case "SMTPSSLEnabled":
@@ -314,7 +330,7 @@ func updateOptionMap(key string, value string) (err error) {
 		common.QuotaForInvitee, _ = strconv.Atoi(value)
 	case "QuotaRemindThreshold":
 		common.QuotaRemindThreshold, _ = strconv.Atoi(value)
-	case "ShouldPreConsumedQuota":
+	case "PreConsumedQuota":
 		common.PreConsumedQuota, _ = strconv.Atoi(value)
 	case "ModelRequestRateLimitCount":
 		setting.ModelRequestRateLimitCount, _ = strconv.Atoi(value)
@@ -329,15 +345,17 @@ 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":
@@ -351,11 +369,34 @@ func updateOptionMap(key string, value string) (err error) {
 	case "SensitiveWords":
 		setting.SensitiveWordsFromString(value)
 	case "AutomaticDisableKeywords":
-		setting.AutomaticDisableKeywordsFromString(value)
-	case "GeminiSafetySettings":
-		setting.GeminiSafetySettingFromJsonString(value)
+		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)

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"
 )
@@ -153,7 +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/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
 }
@@ -64,7 +68,6 @@ func (a *Adaptor) ConvertEmbeddingRequest(c *gin.Context, info *relaycommon.Rela
 	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/dto.go

@@ -14,7 +14,7 @@ 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"`
 }

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/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"
@@ -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/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
 }
 

relay/channel/claude/dto.go

@@ -58,7 +58,7 @@ type ClaudeRequest struct {
 	TopK              int             `json:"top_k,omitempty"`
 	//ClaudeMetadata    `json:"metadata,omitempty"`
 	Stream     bool      `json:"stream,omitempty"`
-	Tools      []Tool    `json:"tools,omitempty"`
+	Tools      any       `json:"tools,omitempty"`
 	ToolChoice any       `json:"tool_choice,omitempty"`
 	Thinking   *Thinking `json:"thinking,omitempty"`
 }

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"
@@ -93,10 +94,12 @@ func RequestOpenAI2ClaudeMessage(textRequest dto.GeneralOpenAIRequest) (*ClaudeR
 		Tools:         claudeTools,
 	}
 
-	if strings.HasSuffix(textRequest.Model, "-thinking") {
-		if claudeRequest.MaxTokens == 0 {
-			claudeRequest.MaxTokens = 8192
-		}
+	if claudeRequest.MaxTokens == 0 {
+		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 {
@@ -106,7 +109,7 @@ func RequestOpenAI2ClaudeMessage(textRequest dto.GeneralOpenAIRequest) (*ClaudeR
 		// BudgetTokens 为 max_tokens 的 80%
 		claudeRequest.Thinking = &Thinking{
 			Type:         "enabled",
-			BudgetTokens: int(float64(claudeRequest.MaxTokens) * 0.8),
+			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
@@ -115,9 +118,6 @@ func RequestOpenAI2ClaudeMessage(textRequest dto.GeneralOpenAIRequest) (*ClaudeR
 		claudeRequest.Model = strings.TrimSuffix(textRequest.Model, "-thinking")
 	}
 
-	if claudeRequest.MaxTokens == 0 {
-		claudeRequest.MaxTokens = 4096
-	}
 	if textRequest.Stop != nil {
 		// stop maybe string/array string, convert to array string
 		switch textRequest.Stop.(type) {
@@ -296,7 +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)
@@ -315,10 +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: "",
 						},
@@ -333,8 +333,8 @@ func StreamResponseClaude2OpenAI(reqMode int, claudeResponse *ClaudeResponse) (*
 				choice.Delta.SetContentString(claudeResponse.Delta.Text)
 				switch claudeResponse.Delta.Type {
 				case "input_json_delta":
-					tools = append(tools, dto.ToolCall{
-						Function: dto.FunctionCall{
+					tools = append(tools, dto.ToolCallResponse{
+						Function: dto.FunctionResponse{
 							Arguments: claudeResponse.Delta.PartialJson,
 						},
 					})
@@ -382,7 +382,7 @@ 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 {
@@ -403,10 +403,10 @@ func ResponseClaude2OpenAI(reqMode int, claudeResponse *ClaudeResponse) *dto.Ope
 			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),
 					},
@@ -443,28 +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
@@ -481,9 +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
@@ -491,11 +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)
@@ -508,14 +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/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/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/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

@@ -7,11 +7,11 @@ import (
 	"io"
 	"net/http"
 	"one-api/common"
-	"one-api/constant"
 	"one-api/dto"
 	"one-api/relay/channel"
 	relaycommon "one-api/relay/common"
 	"one-api/service"
+	"one-api/setting/model_setting"
 
 	"strings"
 
@@ -64,15 +64,7 @@ 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

relay/channel/gemini/constant.go

@@ -22,7 +22,7 @@ var ModelList = []string{
 
 var SafetySettingList = []string{
 	"HARM_CATEGORY_HARASSMENT",
-	"HARM_CATEGORY_VIOLENCE",
+	"HARM_CATEGORY_HATE_SPEECH",
 	"HARM_CATEGORY_SEXUALLY_EXPLICIT",
 	"HARM_CATEGORY_DANGEROUS_CONTENT",
 	"HARM_CATEGORY_CIVIC_INTEGRITY",

relay/channel/gemini/relay-gemini.go

@@ -1,7 +1,6 @@
 package gemini
 
 import (
-	"bufio"
 	"encoding/json"
 	"fmt"
 	"io"
@@ -10,8 +9,9 @@ import (
 	"one-api/constant"
 	"one-api/dto"
 	relaycommon "one-api/relay/common"
+	"one-api/relay/helper"
 	"one-api/service"
-	"one-api/setting"
+	"one-api/setting/model_setting"
 	"strings"
 	"unicode/utf8"
 
@@ -36,14 +36,14 @@ func CovertGemini2OpenAI(textRequest dto.GeneralOpenAIRequest) (*GeminiChatReque
 	for _, category := range SafetySettingList {
 		safetySettings = append(safetySettings, GeminiChatSafetySettings{
 			Category:  category,
-			Threshold: setting.GetGeminiSafetySetting(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 {
@@ -338,7 +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 {
@@ -350,10 +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,
 		},
@@ -368,7 +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),
@@ -380,12 +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 {
@@ -400,9 +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"))
@@ -418,7 +418,7 @@ func responseGeminiChat2OpenAI(response *GeminiChatResponse) *dto.OpenAITextResp
 				choice.FinishReason = constant.FinishReasonContentFilter
 			}
 		}
-		if is_tool_call {
+		if isToolCall {
 			choice.FinishReason = constant.FinishReasonToolCalls
 		}
 
@@ -429,10 +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{
@@ -457,7 +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)
 				}
@@ -482,9 +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) {
@@ -492,27 +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
@@ -521,15 +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
 
@@ -538,14 +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
 }
 

relay/channel/jina/adaptor.go

@@ -61,7 +61,7 @@ func (a *Adaptor) ConvertEmbeddingRequest(c *gin.Context, info *relaycommon.Rela
 
 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/ollama/dto.go

@@ -3,22 +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"`
-	MaxTokens        uint               `json:"max_tokens,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"`
-	Suffix           any                `json:"suffix,omitempty"`
-	StreamOptions    *dto.StreamOptions `json:"stream_options,omitempty"`
-	Prompt           any                `json:"prompt,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 {

relay/channel/openai/adaptor.go

@@ -14,6 +14,7 @@ import (
 	"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"
@@ -146,7 +147,7 @@ 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) {
@@ -228,6 +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

@@ -11,6 +11,7 @@ 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",

relay/channel/openai/relay-openai.go

@@ -1,7 +1,6 @@
 package openai
 
 import (
-	"bufio"
 	"bytes"
 	"encoding/json"
 	"fmt"
@@ -14,11 +13,10 @@ 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"
@@ -32,7 +30,7 @@ func sendStreamData(c *gin.Context, info *relaycommon.RelayInfo, data string, fo
 	}
 
 	if !forceFormat && !thinkToContent {
-		return service.StringData(c, data)
+		return helper.StringData(c, data)
 	}
 
 	var lastStreamResponse dto.ChatCompletionsStreamResponse
@@ -41,44 +39,68 @@ func sendStreamData(c *gin.Context, info *relaycommon.RelayInfo, data string, fo
 	}
 
 	if !thinkToContent {
-		return service.ObjectData(c, lastStreamResponse)
+		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.IsFirstResponse {
-		response := lastStreamResponse.Copy()
-		for i := range response.Choices {
-			response.Choices[i].Delta.SetContentString("<think>\n")
-			response.Choices[i].Delta.SetReasoningContent("")
+	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)
 		}
-		service.ObjectData(c, response)
 	}
 
 	if lastStreamResponse.Choices == nil || len(lastStreamResponse.Choices) == 0 {
-		return service.ObjectData(c, lastStreamResponse)
+		return helper.ObjectData(c, lastStreamResponse)
 	}
 
 	// Process each choice
 	for i, choice := range lastStreamResponse.Choices {
 		// Handle transition from thinking to content
-		if len(choice.Delta.GetContentString()) > 0 && !info.SendLastReasoningResponse {
+		if hasContent && !info.ThinkingContentInfo.SendLastThinkingContent {
 			response := lastStreamResponse.Copy()
 			for j := range response.Choices {
-				response.Choices[j].Delta.SetContentString("\n</think>")
-				response.Choices[j].Delta.SetReasoningContent("")
+				response.Choices[j].Delta.SetContentString("\n</think>\n")
+				response.Choices[j].Delta.ReasoningContent = nil
+				response.Choices[j].Delta.Reasoning = nil
 			}
-			info.SendLastReasoningResponse = true
-			service.ObjectData(c, response)
+			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.SetReasoningContent("")
+			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 service.ObjectData(c, lastStreamResponse)
+	return helper.ObjectData(c, lastStreamResponse)
 }
 
 func OaiStreamHandler(c *gin.Context, resp *http.Response, info *relaycommon.RelayInfo) (*dto.OpenAIErrorWithStatusCode, *dto.Usage) {
@@ -108,64 +130,22 @@ func OaiStreamHandler(c *gin.Context, resp *http.Response, info *relaycommon.Rel
 	}
 
 	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 common.DebugEnabled {
-				println(data)
-			}
-			if len(data) < 6 { // ignore blank line or wrong format
-				continue
-			}
-			if data[:5] != "data:" && data[:6] != "[DONE]" {
-				continue
-			}
-			mu.Lock()
-			data = data[5:]
-			data = strings.TrimSpace(data)
-			if !strings.HasPrefix(data, "[DONE]") {
-				if lastStreamData != "" {
-					err := sendStreamData(c, info, lastStreamData, forceFormat, thinkToContent)
-					if err != nil {
-						common.LogError(c, "streaming error: "+err.Error())
-					}
-					info.SetFirstResponseTime()
-				}
-				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
@@ -210,7 +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)
@@ -231,7 +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())
+					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)
@@ -274,14 +257,14 @@ func OaiStreamHandler(c *gin.Context, resp *http.Response, info *relaycommon.Rel
 	}
 
 	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
 }
 
@@ -323,7 +306,7 @@ 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(choice.Message.StringContent()+choice.Message.ReasoningContent, model)
+			ctkm, _ := service.CountTextToken(choice.Message.StringContent()+choice.Message.ReasoningContent+choice.Message.Reasoning, model)
 			completionTokens += ctkm
 		}
 		simpleResponse.Usage = dto.Usage{
@@ -512,7 +495,7 @@ 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
@@ -618,7 +601,7 @@ 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/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/tencent/relay-tencent.go

@@ -14,6 +14,7 @@ import (
 	"one-api/common"
 	"one-api/constant"
 	"one-api/dto"
+	"one-api/relay/helper"
 	"one-api/service"
 	"strconv"
 	"strings"
@@ -91,7 +92,7 @@ func tencentStreamHandler(c *gin.Context, resp *http.Response) (*dto.OpenAIError
 	scanner := bufio.NewScanner(resp.Body)
 	scanner.Split(bufio.ScanLines)
 
-	service.SetEventStreamHeaders(c)
+	helper.SetEventStreamHeaders(c)
 
 	for scanner.Scan() {
 		data := scanner.Text()
@@ -112,7 +113,7 @@ func tencentStreamHandler(c *gin.Context, resp *http.Response) (*dto.OpenAIError
 			responseText += response.Choices[0].Delta.GetContentString()
 		}
 
-		err = service.ObjectData(c, response)
+		err = helper.ObjectData(c, response)
 		if err != nil {
 			common.SysError(err.Error())
 		}
@@ -122,7 +123,7 @@ func tencentStreamHandler(c *gin.Context, resp *http.Response) (*dto.OpenAIError
 		common.SysError("error reading stream: " + err.Error())
 	}
 
-	service.Done(c)
+	helper.Done(c)
 
 	err := resp.Body.Close()
 	if err != nil {

relay/channel/vertex/adaptor.go

@@ -5,7 +5,6 @@ import (
 	"errors"
 	"fmt"
 	"github.com/gin-gonic/gin"
-	"github.com/jinzhu/copier"
 	"io"
 	"net/http"
 	"one-api/dto"
@@ -28,6 +27,8 @@ var claudeModelMap = map[string]string{
 	"claude-3-opus-20240229":     "claude-3-opus@20240229",
 	"claude-3-haiku-20240307":    "claude-3-haiku@20240307",
 	"claude-3-5-sonnet-20240620": "claude-3-5-sonnet@20240620",
+	"claude-3-5-sonnet-20241022": "claude-3-5-sonnet-v2@20241022",
+	"claude-3-7-sonnet-20250219": "claude-3-7-sonnet@20250219",
 }
 
 const anthropicVersion = "vertex-2023-10-16"
@@ -85,15 +86,16 @@ func (a *Adaptor) GetRequestURL(info *relaycommon.RelayInfo) (string, error) {
 		} else {
 			suffix = "rawPredict"
 		}
+		model := info.UpstreamModelName
 		if v, ok := claudeModelMap[info.UpstreamModelName]; ok {
-			info.UpstreamModelName = v
+			model = v
 		}
 		return fmt.Sprintf(
 			"https://%s-aiplatform.googleapis.com/v1/projects/%s/locations/%s/publishers/anthropic/models/%s:%s",
 			region,
 			adc.ProjectID,
 			region,
-			info.UpstreamModelName,
+			model,
 			suffix,
 		), nil
 	} else if a.RequestMode == RequestModeLlama {
@@ -126,13 +128,9 @@ func (a *Adaptor) ConvertRequest(c *gin.Context, info *relaycommon.RelayInfo, re
 		if err != nil {
 			return nil, err
 		}
-		vertexClaudeReq := &VertexAIClaudeRequest{
-			AnthropicVersion: anthropicVersion,
-		}
-		if err = copier.Copy(vertexClaudeReq, claudeReq); err != nil {
-			return nil, errors.New("failed to copy claude request")
-		}
-		c.Set("request_model", request.Model)
+		vertexClaudeReq := copyRequest(claudeReq, anthropicVersion)
+		c.Set("request_model", claudeReq.Model)
+		info.UpstreamModelName = claudeReq.Model
 		return vertexClaudeReq, nil
 	} else if a.RequestMode == RequestModeGemini {
 		geminiRequest, err := gemini.CovertGemini2OpenAI(*request)
@@ -156,7 +154,6 @@ func (a *Adaptor) ConvertEmbeddingRequest(c *gin.Context, info *relaycommon.Rela
 	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/vertex/dto.go

@@ -1,17 +1,37 @@
 package vertex
 
-import "one-api/relay/channel/claude"
+import (
+	"one-api/relay/channel/claude"
+)
 
 type VertexAIClaudeRequest struct {
 	AnthropicVersion string                 `json:"anthropic_version"`
 	Messages         []claude.ClaudeMessage `json:"messages"`
-	System           string                 `json:"system,omitempty"`
-	MaxTokens        int                    `json:"max_tokens,omitempty"`
+	System           any                    `json:"system,omitempty"`
+	MaxTokens        uint                   `json:"max_tokens,omitempty"`
 	StopSequences    []string               `json:"stop_sequences,omitempty"`
 	Stream           bool                   `json:"stream,omitempty"`
 	Temperature      *float64               `json:"temperature,omitempty"`
 	TopP             float64                `json:"top_p,omitempty"`
 	TopK             int                    `json:"top_k,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, version string) *VertexAIClaudeRequest {
+	return &VertexAIClaudeRequest{
+		AnthropicVersion: version,
+		System:           req.System,
+		Messages:         req.Messages,
+		MaxTokens:        req.MaxTokens,
+		Stream:           req.Stream,
+		Temperature:      req.Temperature,
+		TopP:             req.TopP,
+		TopK:             req.TopK,
+		StopSequences:    req.StopSequences,
+		Tools:            req.Tools,
+		ToolChoice:       req.ToolChoice,
+		Thinking:         req.Thinking,
+	}
 }

relay/channel/xunfei/relay-xunfei.go

@@ -14,6 +14,7 @@ import (
 	"one-api/common"
 	"one-api/constant"
 	"one-api/dto"
+	"one-api/relay/helper"
 	"one-api/service"
 	"strings"
 	"time"
@@ -132,7 +133,7 @@ func xunfeiStreamHandler(c *gin.Context, textRequest dto.GeneralOpenAIRequest, a
 	if err != nil {
 		return service.OpenAIErrorWrapper(err, "make xunfei request err", http.StatusInternalServerError), nil
 	}
-	service.SetEventStreamHeaders(c)
+	helper.SetEventStreamHeaders(c)
 	var usage dto.Usage
 	c.Stream(func(w io.Writer) bool {
 		select {

relay/channel/zhipu/relay-zhipu.go

@@ -10,6 +10,7 @@ import (
 	"one-api/common"
 	"one-api/constant"
 	"one-api/dto"
+	"one-api/relay/helper"
 	"one-api/service"
 	"strings"
 	"sync"
@@ -177,7 +178,7 @@ func zhipuStreamHandler(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/zhipu_4v/relay-zhipu_v4.go

@@ -10,6 +10,7 @@ import (
 	"net/http"
 	"one-api/common"
 	"one-api/dto"
+	"one-api/relay/helper"
 	"one-api/service"
 	"strings"
 	"sync"
@@ -197,7 +198,7 @@ func zhipuStreamHandler(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/common/relay_info.go

@@ -12,25 +12,30 @@ import (
 	"github.com/gorilla/websocket"
 )
 
+type ThinkingContentInfo struct {
+	IsFirstThinkingContent  bool
+	SendLastThinkingContent bool
+}
+
 type RelayInfo struct {
-	ChannelType               int
-	ChannelId                 int
-	TokenId                   int
-	TokenKey                  string
-	UserId                    int
-	Group                     string
-	TokenUnlimited            bool
-	StartTime                 time.Time
-	FirstResponseTime         time.Time
-	IsFirstResponse           bool
-	SendLastReasoningResponse bool
-	ApiType                   int
-	IsStream                  bool
-	IsPlayground              bool
-	UsePrice                  bool
-	RelayMode                 int
-	UpstreamModelName         string
-	OriginModelName           string
+	ChannelType       int
+	ChannelId         int
+	TokenId           int
+	TokenKey          string
+	UserId            int
+	Group             string
+	TokenUnlimited    bool
+	StartTime         time.Time
+	FirstResponseTime time.Time
+	isFirstResponse   bool
+	//SendLastReasoningResponse bool
+	ApiType           int
+	IsStream          bool
+	IsPlayground      bool
+	UsePrice          bool
+	RelayMode         int
+	UpstreamModelName string
+	OriginModelName   string
 	//RecodeModelName      string
 	RequestURLPath       string
 	ApiVersion           string
@@ -53,6 +58,7 @@ type RelayInfo struct {
 	UserSetting          map[string]interface{}
 	UserEmail            string
 	UserQuota            int
+	ThinkingContentInfo
 }
 
 // 定义支持流式选项的通道类型
@@ -95,7 +101,7 @@ func GenRelayInfo(c *gin.Context) *RelayInfo {
 		UserQuota:         c.GetInt(constant.ContextKeyUserQuota),
 		UserSetting:       c.GetStringMap(constant.ContextKeyUserSetting),
 		UserEmail:         c.GetString(constant.ContextKeyUserEmail),
-		IsFirstResponse:   true,
+		isFirstResponse:   true,
 		RelayMode:         relayconstant.Path2RelayMode(c.Request.URL.Path),
 		BaseUrl:           c.GetString("base_url"),
 		RequestURLPath:    c.Request.URL.String(),
@@ -117,6 +123,10 @@ func GenRelayInfo(c *gin.Context) *RelayInfo {
 		ApiKey:         strings.TrimPrefix(c.Request.Header.Get("Authorization"), "Bearer "),
 		Organization:   c.GetString("channel_organization"),
 		ChannelSetting: channelSetting,
+		ThinkingContentInfo: ThinkingContentInfo{
+			IsFirstThinkingContent:  true,
+			SendLastThinkingContent: false,
+		},
 	}
 	if strings.HasPrefix(c.Request.URL.Path, "/pg") {
 		info.IsPlayground = true
@@ -147,9 +157,9 @@ func (info *RelayInfo) SetIsStream(isStream bool) {
 }
 
 func (info *RelayInfo) SetFirstResponseTime() {
-	if info.IsFirstResponse {
+	if info.isFirstResponse {
 		info.FirstResponseTime = time.Now()
-		info.IsFirstResponse = false
+		info.isFirstResponse = false
 	}
 }
 

relay/constant/api_type.go

@@ -30,6 +30,7 @@ const (
 	APITypeMokaAI
 	APITypeVolcEngine
 	APITypeBaiduV2
+	APITypeOpenRouter
 	APITypeDummy // this one is only for count, do not add any channel after this
 )
 
@@ -86,6 +87,8 @@ func ChannelType2APIType(channelType int) (int, bool) {
 		apiType = APITypeVolcEngine
 	case common.ChannelTypeBaiduV2:
 		apiType = APITypeBaiduV2
+	case common.ChannelTypeOpenRouter:
+		apiType = APITypeOpenRouter
 	}
 	if apiType == -1 {
 		return APITypeOpenAI, false

relay/helper/common.go

@@ -1,4 +1,4 @@
-package service
+package helper
 
 import (
 	"encoding/json"

relay/helper/price.go

@@ -1,31 +1,47 @@
 package helper
 
 import (
+	"fmt"
 	"github.com/gin-gonic/gin"
 	"one-api/common"
 	relaycommon "one-api/relay/common"
 	"one-api/setting"
+	"one-api/setting/operation_setting"
 )
 
 type PriceData struct {
 	ModelPrice             float64
 	ModelRatio             float64
+	CompletionRatio        float64
+	CacheRatio             float64
 	GroupRatio             float64
 	UsePrice               bool
 	ShouldPreConsumedQuota int
 }
 
-func ModelPriceHelper(c *gin.Context, info *relaycommon.RelayInfo, promptTokens int, maxTokens int) PriceData {
-	modelPrice, usePrice := common.GetModelPrice(info.OriginModelName, false)
+func ModelPriceHelper(c *gin.Context, info *relaycommon.RelayInfo, promptTokens int, maxTokens int) (PriceData, error) {
+	modelPrice, usePrice := operation_setting.GetModelPrice(info.OriginModelName, false)
 	groupRatio := setting.GetGroupRatio(info.Group)
 	var preConsumedQuota int
 	var modelRatio float64
+	var completionRatio float64
+	var cacheRatio float64
 	if !usePrice {
 		preConsumedTokens := common.PreConsumedQuota
 		if maxTokens != 0 {
 			preConsumedTokens = promptTokens + maxTokens
 		}
-		modelRatio = common.GetModelRatio(info.OriginModelName)
+		var success bool
+		modelRatio, success = operation_setting.GetModelRatio(info.OriginModelName)
+		if !success {
+			if info.UserId == 1 {
+				return PriceData{}, fmt.Errorf("模型 %s 倍率或价格未配置，请设置或开始自用模式；Model %s ratio or price not set, please set or start self-use mode", info.OriginModelName, info.OriginModelName)
+			} else {
+				return PriceData{}, fmt.Errorf("模型 %s 倍率或价格未配置, 请联系管理员设置；Model %s ratio or price not set, please contact administrator to set", info.OriginModelName, info.OriginModelName)
+			}
+		}
+		completionRatio = operation_setting.GetCompletionRatio(info.OriginModelName)
+		cacheRatio, _ = operation_setting.GetCacheRatio(info.OriginModelName)
 		ratio := modelRatio * groupRatio
 		preConsumedQuota = int(float64(preConsumedTokens) * ratio)
 	} else {
@@ -34,8 +50,10 @@ func ModelPriceHelper(c *gin.Context, info *relaycommon.RelayInfo, promptTokens
 	return PriceData{
 		ModelPrice:             modelPrice,
 		ModelRatio:             modelRatio,
+		CompletionRatio:        completionRatio,
 		GroupRatio:             groupRatio,
 		UsePrice:               usePrice,
+		CacheRatio:             cacheRatio,
 		ShouldPreConsumedQuota: preConsumedQuota,
-	}
+	}, nil
 }

relay/helper/stream_scanner.go

@@ -0,0 +1,91 @@
+package helper
+
+import (
+	"bufio"
+	"context"
+	"io"
+	"net/http"
+	"one-api/common"
+	"one-api/constant"
+	relaycommon "one-api/relay/common"
+	"strings"
+	"time"
+
+	"github.com/gin-gonic/gin"
+)
+
+func StreamScannerHandler(c *gin.Context, resp *http.Response, info *relaycommon.RelayInfo, dataHandler func(data string) bool) {
+
+	if resp == nil {
+		return
+	}
+
+	defer resp.Body.Close()
+
+	streamingTimeout := time.Duration(constant.StreamingTimeout) * time.Second
+	if strings.HasPrefix(info.UpstreamModelName, "o1") || strings.HasPrefix(info.UpstreamModelName, "o3") {
+		// twice timeout for thinking model
+		streamingTimeout *= 2
+	}
+
+	var (
+		stopChan = make(chan bool, 2)
+		scanner  = bufio.NewScanner(resp.Body)
+		ticker   = time.NewTicker(streamingTimeout)
+	)
+
+	defer func() {
+		ticker.Stop()
+		close(stopChan)
+	}()
+
+	scanner.Split(bufio.ScanLines)
+	SetEventStreamHeaders(c)
+
+	ctx, cancel := context.WithCancel(context.Background())
+	defer cancel()
+
+	ctx = context.WithValue(ctx, "stop_chan", stopChan)
+	common.RelayCtxGo(ctx, func() {
+		for scanner.Scan() {
+			ticker.Reset(streamingTimeout)
+			data := scanner.Text()
+			if common.DebugEnabled {
+				println(data)
+			}
+
+			if len(data) < 6 {
+				continue
+			}
+			if data[:5] != "data:" && data[:6] != "[DONE]" {
+				continue
+			}
+			data = data[5:]
+			data = strings.TrimLeft(data, " ")
+			data = strings.TrimSuffix(data, "\"")
+			if !strings.HasPrefix(data, "[DONE]") {
+				info.SetFirstResponseTime()
+				success := dataHandler(data)
+				if !success {
+					break
+				}
+			}
+		}
+
+		if err := scanner.Err(); err != nil {
+			if err != io.EOF {
+				common.LogError(c, "scanner error: "+err.Error())
+			}
+		}
+
+		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
+	}
+}

relay/relay-audio.go

@@ -7,7 +7,6 @@ import (
 	"net/http"
 	"one-api/common"
 	"one-api/dto"
-	"one-api/model"
 	relaycommon "one-api/relay/common"
 	relayconstant "one-api/relay/constant"
 	"one-api/relay/helper"
@@ -75,12 +74,11 @@ func AudioHelper(c *gin.Context) (openaiErr *dto.OpenAIErrorWithStatusCode) {
 		relayInfo.PromptTokens = promptTokens
 	}
 
-	priceData := helper.ModelPriceHelper(c, relayInfo, preConsumedTokens, 0)
-
-	userQuota, err := model.GetUserQuota(relayInfo.UserId, false)
+	priceData, err := helper.ModelPriceHelper(c, relayInfo, preConsumedTokens, 0)
 	if err != nil {
-		return service.OpenAIErrorWrapperLocal(err, "get_user_quota_failed", http.StatusInternalServerError)
+		return service.OpenAIErrorWrapperLocal(err, "model_price_error", http.StatusInternalServerError)
 	}
+
 	preConsumedQuota, userQuota, openaiErr := preConsumeQuota(c, priceData.ShouldPreConsumedQuota, relayInfo)
 	if openaiErr != nil {
 		return openaiErr

relay/relay-image.go

@@ -86,7 +86,10 @@ func ImageHelper(c *gin.Context) *dto.OpenAIErrorWithStatusCode {
 
 	imageRequest.Model = relayInfo.UpstreamModelName
 
-	priceData := helper.ModelPriceHelper(c, relayInfo, 0, 0)
+	priceData, err := helper.ModelPriceHelper(c, relayInfo, 0, 0)
+	if err != nil {
+		return service.OpenAIErrorWrapperLocal(err, "model_price_error", http.StatusInternalServerError)
+	}
 	if !priceData.UsePrice {
 		// modelRatio 16 = modelPrice $0.04
 		// per 1 modelRatio = $0.04 / 16

relay/relay-mj.go

@@ -15,6 +15,7 @@ import (
 	relayconstant "one-api/relay/constant"
 	"one-api/service"
 	"one-api/setting"
+	"one-api/setting/operation_setting"
 	"strconv"
 	"strings"
 	"time"
@@ -157,10 +158,10 @@ func RelaySwapFace(c *gin.Context) *dto.MidjourneyResponse {
 		return service.MidjourneyErrorWrapper(constant.MjRequestError, "sour_base64_and_target_base64_is_required")
 	}
 	modelName := service.CoverActionToModelName(constant.MjActionSwapFace)
-	modelPrice, success := common.GetModelPrice(modelName, true)
+	modelPrice, success := operation_setting.GetModelPrice(modelName, true)
 	// 如果没有配置价格，则使用默认价格
 	if !success {
-		defaultPrice, ok := common.GetDefaultModelRatioMap()[modelName]
+		defaultPrice, ok := operation_setting.GetDefaultModelRatioMap()[modelName]
 		if !ok {
 			modelPrice = 0.1
 		} else {
@@ -463,10 +464,10 @@ func RelayMidjourneySubmit(c *gin.Context, relayMode int) *dto.MidjourneyRespons
 	fullRequestURL := fmt.Sprintf("%s%s", baseURL, requestURL)
 
 	modelName := service.CoverActionToModelName(midjRequest.Action)
-	modelPrice, success := common.GetModelPrice(modelName, true)
+	modelPrice, success := operation_setting.GetModelPrice(modelName, true)
 	// 如果没有配置价格，则使用默认价格
 	if !success {
-		defaultPrice, ok := common.GetDefaultModelRatioMap()[modelName]
+		defaultPrice, ok := operation_setting.GetDefaultModelRatioMap()[modelName]
 		if !ok {
 			modelPrice = 0.1
 		} else {

relay/relay-text.go

@@ -106,7 +106,10 @@ func TextHelper(c *gin.Context) (openaiErr *dto.OpenAIErrorWithStatusCode) {
 		c.Set("prompt_tokens", promptTokens)
 	}
 
-	priceData := helper.ModelPriceHelper(c, relayInfo, promptTokens, int(textRequest.MaxTokens))
+	priceData, err := helper.ModelPriceHelper(c, relayInfo, promptTokens, int(textRequest.MaxTokens))
+	if err != nil {
+		return service.OpenAIErrorWrapperLocal(err, "model_price_error", http.StatusInternalServerError)
+	}
 
 	// pre-consume quota 预消耗配额
 	preConsumedQuota, userQuota, openaiErr := preConsumeQuota(c, priceData.ShouldPreConsumedQuota, relayInfo)
@@ -301,24 +304,26 @@ func postConsumeQuota(ctx *gin.Context, relayInfo *relaycommon.RelayInfo,
 			CompletionTokens: 0,
 			TotalTokens:      relayInfo.PromptTokens,
 		}
-		extraContent += "  ，（可能是请求出错）"
+		extraContent += "（可能是请求出错）"
 	}
 	useTimeSeconds := time.Now().Unix() - relayInfo.StartTime.Unix()
 	promptTokens := usage.PromptTokens
+	cacheTokens := usage.PromptTokensDetails.CachedTokens
 	completionTokens := usage.CompletionTokens
 	modelName := relayInfo.OriginModelName
 
 	tokenName := ctx.GetString("token_name")
-	completionRatio := common.GetCompletionRatio(modelName)
+	completionRatio := priceData.CompletionRatio
+	cacheRatio := priceData.CacheRatio
 	ratio := priceData.ModelRatio * priceData.GroupRatio
 	modelRatio := priceData.ModelRatio
 	groupRatio := priceData.GroupRatio
 	modelPrice := priceData.ModelPrice
-	usePrice := priceData.UsePrice
 
 	quota := 0
 	if !priceData.UsePrice {
-		quota = promptTokens + int(math.Round(float64(completionTokens)*completionRatio))
+		quota = (promptTokens - cacheTokens) + int(math.Round(float64(cacheTokens)*cacheRatio))
+		quota += int(math.Round(float64(completionTokens) * completionRatio))
 		quota = int(math.Round(float64(quota) * ratio))
 		if ratio != 0 && quota <= 0 {
 			quota = 1
@@ -327,8 +332,9 @@ func postConsumeQuota(ctx *gin.Context, relayInfo *relaycommon.RelayInfo,
 		quota = int(modelPrice * common.QuotaPerUnit * groupRatio)
 	}
 	totalTokens := promptTokens + completionTokens
+	
 	var logContent string
-	if !usePrice {
+	if !priceData.UsePrice {
 		logContent = fmt.Sprintf("模型倍率 %.2f，补全倍率 %.2f，分组倍率 %.2f", modelRatio, completionRatio, groupRatio)
 	} else {
 		logContent = fmt.Sprintf("模型价格 %.2f，分组倍率 %.2f", modelPrice, groupRatio)
@@ -369,7 +375,7 @@ func postConsumeQuota(ctx *gin.Context, relayInfo *relaycommon.RelayInfo,
 	if extraContent != "" {
 		logContent += ", " + extraContent
 	}
-	other := service.GenerateTextOtherInfo(ctx, relayInfo, modelRatio, groupRatio, completionRatio, modelPrice)
+	other := service.GenerateTextOtherInfo(ctx, relayInfo, modelRatio, groupRatio, completionRatio, cacheTokens, cacheRatio, modelPrice)
 	model.RecordConsumeLog(ctx, relayInfo.UserId, relayInfo.ChannelId, promptTokens, completionTokens, logModel,
 		tokenName, quota, logContent, relayInfo.TokenId, userQuota, int(useTimeSeconds), relayInfo.IsStream, relayInfo.Group, other)
 

relay/relay_adaptor.go

@@ -18,6 +18,7 @@ import (
 	"one-api/relay/channel/mokaai"
 	"one-api/relay/channel/ollama"
 	"one-api/relay/channel/openai"
+	"one-api/relay/channel/openrouter"
 	"one-api/relay/channel/palm"
 	"one-api/relay/channel/perplexity"
 	"one-api/relay/channel/siliconflow"
@@ -83,6 +84,8 @@ func GetAdaptor(apiType int) channel.Adaptor {
 		return &volcengine.Adaptor{}
 	case constant.APITypeBaiduV2:
 		return &baidu_v2.Adaptor{}
+	case constant.APITypeOpenRouter:
+		return &openrouter.Adaptor{}
 	}
 	return nil
 }

relay/relay_embedding.go

@@ -57,8 +57,10 @@ func EmbeddingHelper(c *gin.Context) (openaiErr *dto.OpenAIErrorWithStatusCode)
 	promptToken := getEmbeddingPromptToken(*embeddingRequest)
 	relayInfo.PromptTokens = promptToken
 
-	priceData := helper.ModelPriceHelper(c, relayInfo, promptToken, 0)
-
+	priceData, err := helper.ModelPriceHelper(c, relayInfo, promptToken, 0)
+	if err != nil {
+		return service.OpenAIErrorWrapperLocal(err, "model_price_error", http.StatusInternalServerError)
+	}
 	// pre-consume quota 预消耗配额
 	preConsumedQuota, userQuota, openaiErr := preConsumeQuota(c, priceData.ShouldPreConsumedQuota, relayInfo)
 	if openaiErr != nil {

relay/relay_rerank.go

@@ -50,8 +50,10 @@ func RerankHelper(c *gin.Context, relayMode int) (openaiErr *dto.OpenAIErrorWith
 	promptToken := getRerankPromptToken(*rerankRequest)
 	relayInfo.PromptTokens = promptToken
 
-	priceData := helper.ModelPriceHelper(c, relayInfo, promptToken, 0)
-
+	priceData, err := helper.ModelPriceHelper(c, relayInfo, promptToken, 0)
+	if err != nil {
+		return service.OpenAIErrorWrapperLocal(err, "model_price_error", http.StatusInternalServerError)
+	}
 	// pre-consume quota 预消耗配额
 	preConsumedQuota, userQuota, openaiErr := preConsumeQuota(c, priceData.ShouldPreConsumedQuota, relayInfo)
 	if openaiErr != nil {

relay/relay_task.go

@@ -16,6 +16,7 @@ import (
 	relayconstant "one-api/relay/constant"
 	"one-api/service"
 	"one-api/setting"
+	"one-api/setting/operation_setting"
 )
 
 /*
@@ -37,9 +38,9 @@ func RelayTaskSubmit(c *gin.Context, relayMode int) (taskErr *dto.TaskError) {
 	}
 
 	modelName := service.CoverTaskActionToModelName(platform, relayInfo.Action)
-	modelPrice, success := common.GetModelPrice(modelName, true)
+	modelPrice, success := operation_setting.GetModelPrice(modelName, true)
 	if !success {
-		defaultPrice, ok := common.GetDefaultModelRatioMap()[modelName]
+		defaultPrice, ok := operation_setting.GetDefaultModelRatioMap()[modelName]
 		if !ok {
 			modelPrice = 0.1
 		} else {

relay/websocket.go

@@ -11,6 +11,7 @@ import (
 	relaycommon "one-api/relay/common"
 	"one-api/service"
 	"one-api/setting"
+	"one-api/setting/operation_setting"
 )
 
 func WssHelper(c *gin.Context, ws *websocket.Conn) (openaiErr *dto.OpenAIErrorWithStatusCode) {
@@ -39,7 +40,7 @@ func WssHelper(c *gin.Context, ws *websocket.Conn) (openaiErr *dto.OpenAIErrorWi
 		}
 	}
 	//relayInfo.UpstreamModelName = textRequest.Model
-	modelPrice, getModelPriceSuccess := common.GetModelPrice(relayInfo.UpstreamModelName, false)
+	modelPrice, getModelPriceSuccess := operation_setting.GetModelPrice(relayInfo.UpstreamModelName, false)
 	groupRatio := setting.GetGroupRatio(relayInfo.Group)
 
 	var preConsumedQuota int
@@ -65,7 +66,7 @@ func WssHelper(c *gin.Context, ws *websocket.Conn) (openaiErr *dto.OpenAIErrorWi
 		//if realtimeEvent.Session.MaxResponseOutputTokens != 0 {
 		//	preConsumedTokens = promptTokens + int(realtimeEvent.Session.MaxResponseOutputTokens)
 		//}
-		modelRatio = common.GetModelRatio(relayInfo.UpstreamModelName)
+		modelRatio, _ = operation_setting.GetModelRatio(relayInfo.UpstreamModelName)
 		ratio = modelRatio * groupRatio
 		preConsumedQuota = int(float64(preConsumedTokens) * ratio)
 	} else {

router/api-router.go

@@ -84,6 +84,7 @@ func SetApiRouter(router *gin.Engine) {
 			channelRoute.GET("/", controller.GetAllChannels)
 			channelRoute.GET("/search", controller.SearchChannels)
 			channelRoute.GET("/models", controller.ChannelListModels)
+			channelRoute.GET("/models_enabled", controller.EnabledListModels)
 			channelRoute.GET("/:id", controller.GetChannel)
 			channelRoute.GET("/test", controller.TestAllChannels)
 			channelRoute.GET("/test/:id", controller.TestChannel)

service/channel.go

@@ -6,23 +6,31 @@ import (
 	"one-api/common"
 	"one-api/dto"
 	"one-api/model"
-	"one-api/setting"
+	"one-api/setting/operation_setting"
 	"strings"
 )
 
+func formatNotifyType(channelId int, status int) string {
+	return fmt.Sprintf("%s_%d_%d", dto.NotifyTypeChannelUpdate, channelId, status)
+}
+
 // disable & notify
 func DisableChannel(channelId int, channelName string, reason string) {
-	model.UpdateChannelStatusById(channelId, common.ChannelStatusAutoDisabled, reason)
-	subject := fmt.Sprintf("通道「%s」（#%d）已被禁用", channelName, channelId)
-	content := fmt.Sprintf("通道「%s」（#%d）已被禁用，原因：%s", channelName, channelId, reason)
-	NotifyRootUser(subject, content, dto.NotifyTypeChannelUpdate)
+	success := model.UpdateChannelStatusById(channelId, common.ChannelStatusAutoDisabled, reason)
+	if success {
+		subject := fmt.Sprintf("通道「%s」（#%d）已被禁用", channelName, channelId)
+		content := fmt.Sprintf("通道「%s」（#%d）已被禁用，原因：%s", channelName, channelId, reason)
+		NotifyRootUser(formatNotifyType(channelId, common.ChannelStatusAutoDisabled), subject, content)
+	}
 }
 
 func EnableChannel(channelId int, channelName string) {
-	model.UpdateChannelStatusById(channelId, common.ChannelStatusEnabled, "")
-	subject := fmt.Sprintf("通道「%s」（#%d）已被启用", channelName, channelId)
-	content := fmt.Sprintf("通道「%s」（#%d）已被启用", channelName, channelId)
-	NotifyRootUser(subject, content, dto.NotifyTypeChannelUpdate)
+	success := model.UpdateChannelStatusById(channelId, common.ChannelStatusEnabled, "")
+	if success {
+		subject := fmt.Sprintf("通道「%s」（#%d）已被启用", channelName, channelId)
+		content := fmt.Sprintf("通道「%s」（#%d）已被启用", channelName, channelId)
+		NotifyRootUser(formatNotifyType(channelId, common.ChannelStatusEnabled), subject, content)
+	}
 }
 
 func ShouldDisableChannel(channelType int, err *dto.OpenAIErrorWithStatusCode) bool {
@@ -67,7 +75,7 @@ func ShouldDisableChannel(channelType int, err *dto.OpenAIErrorWithStatusCode) b
 	}
 
 	lowerMessage := strings.ToLower(err.Error.Message)
-	search, _ := AcSearch(lowerMessage, setting.AutomaticDisableKeywords, true)
+	search, _ := AcSearch(lowerMessage, operation_setting.AutomaticDisableKeywords, true)
 	if search {
 		return true
 	}

service/image.go

@@ -7,7 +7,9 @@ import (
 	"fmt"
 	"image"
 	"io"
+	"net/http"
 	"one-api/common"
+	"one-api/constant"
 	"strings"
 
 	"golang.org/x/image/webp"
@@ -23,7 +25,7 @@ func DecodeBase64ImageData(base64String string) (image.Config, string, string, e
 	decodedData, err := base64.StdEncoding.DecodeString(base64String)
 	if err != nil {
 		fmt.Println("Error: Failed to decode base64 string")
-		return image.Config{}, "", "", err
+		return image.Config{}, "", "", fmt.Errorf("failed to decode base64 string: %s", err.Error())
 	}
 
 	// 创建一个bytes.Buffer用于存储解码后的数据
@@ -61,20 +63,51 @@ func DecodeBase64FileData(base64String string) (string, string, error) {
 func GetImageFromUrl(url string) (mimeType string, data string, err error) {
 	resp, err := DoDownloadRequest(url)
 	if err != nil {
-		return "", "", err
-	}
-	if !strings.HasPrefix(resp.Header.Get("Content-Type"), "image/") {
-		return "", "", fmt.Errorf("invalid content type: %s, required image/*", resp.Header.Get("Content-Type"))
+		return "", "", fmt.Errorf("failed to download image: %w", err)
 	}
 	defer resp.Body.Close()
-	buffer := bytes.NewBuffer(nil)
-	_, err = buffer.ReadFrom(resp.Body)
-	if err != nil {
-		return
+
+	// Check HTTP status code
+	if resp.StatusCode != http.StatusOK {
+		return "", "", fmt.Errorf("failed to download image: HTTP %d", resp.StatusCode)
 	}
-	mimeType = resp.Header.Get("Content-Type")
+
+	contentType := resp.Header.Get("Content-Type")
+	if contentType != "application/octet-stream" && !strings.HasPrefix(contentType, "image/") {
+		return "", "", fmt.Errorf("invalid content type: %s, required image/*", contentType)
+	}
+	maxImageSize := int64(constant.MaxFileDownloadMB * 1024 * 1024)
+
+	// Check Content-Length if available
+	if resp.ContentLength > maxImageSize {
+		return "", "", fmt.Errorf("image size %d exceeds maximum allowed size of %d bytes", resp.ContentLength, maxImageSize)
+	}
+
+	// Use LimitReader to prevent reading oversized images
+	limitReader := io.LimitReader(resp.Body, maxImageSize)
+	buffer := &bytes.Buffer{}
+
+	written, err := io.Copy(buffer, limitReader)
+	if err != nil {
+		return "", "", fmt.Errorf("failed to read image data: %w", err)
+	}
+	if written >= maxImageSize {
+		return "", "", fmt.Errorf("image size exceeds maximum allowed size of %d bytes", maxImageSize)
+	}
+
 	data = base64.StdEncoding.EncodeToString(buffer.Bytes())
-	return
+	mimeType = contentType
+
+	// Handle application/octet-stream type
+	if mimeType == "application/octet-stream" {
+		_, format, _, err := DecodeBase64ImageData(data)
+		if err != nil {
+			return "", "", err
+		}
+		mimeType = "image/" + format
+	}
+
+	return mimeType, data, nil
 }
 
 func DecodeUrlImageData(imageUrl string) (image.Config, string, error) {
@@ -92,7 +125,7 @@ func DecodeUrlImageData(imageUrl string) (image.Config, string, error) {
 
 	mimeType := response.Header.Get("Content-Type")
 
-	if !strings.HasPrefix(mimeType, "image/") {
+	if mimeType != "application/octet-stream" && !strings.HasPrefix(mimeType, "image/") {
 		return image.Config{}, "", fmt.Errorf("invalid content type: %s, required image/*", mimeType)
 	}
 

service/log_info_generate.go

@@ -1,16 +1,20 @@
 package service
 
 import (
-	"github.com/gin-gonic/gin"
 	"one-api/dto"
 	relaycommon "one-api/relay/common"
+
+	"github.com/gin-gonic/gin"
 )
 
-func GenerateTextOtherInfo(ctx *gin.Context, relayInfo *relaycommon.RelayInfo, modelRatio, groupRatio, completionRatio, modelPrice float64) map[string]interface{} {
+func GenerateTextOtherInfo(ctx *gin.Context, relayInfo *relaycommon.RelayInfo, modelRatio, groupRatio, completionRatio float64,
+	cacheTokens int, cacheRatio float64, modelPrice float64) map[string]interface{} {
 	other := make(map[string]interface{})
 	other["model_ratio"] = modelRatio
 	other["group_ratio"] = groupRatio
 	other["completion_ratio"] = completionRatio
+	other["cache_tokens"] = cacheTokens
+	other["cache_ratio"] = cacheRatio
 	other["model_price"] = modelPrice
 	other["frt"] = float64(relayInfo.FirstResponseTime.UnixMilli() - relayInfo.StartTime.UnixMilli())
 	if relayInfo.ReasoningEffort != "" {
@@ -27,7 +31,7 @@ func GenerateTextOtherInfo(ctx *gin.Context, relayInfo *relaycommon.RelayInfo, m
 }
 
 func GenerateWssOtherInfo(ctx *gin.Context, relayInfo *relaycommon.RelayInfo, usage *dto.RealtimeUsage, modelRatio, groupRatio, completionRatio, audioRatio, audioCompletionRatio, modelPrice float64) map[string]interface{} {
-	info := GenerateTextOtherInfo(ctx, relayInfo, modelRatio, groupRatio, completionRatio, modelPrice)
+	info := GenerateTextOtherInfo(ctx, relayInfo, modelRatio, groupRatio, completionRatio, 0, 0.0, modelPrice)
 	info["ws"] = true
 	info["audio_input"] = usage.InputTokenDetails.AudioTokens
 	info["audio_output"] = usage.OutputTokenDetails.AudioTokens
@@ -39,7 +43,7 @@ func GenerateWssOtherInfo(ctx *gin.Context, relayInfo *relaycommon.RelayInfo, us
 }
 
 func GenerateAudioOtherInfo(ctx *gin.Context, relayInfo *relaycommon.RelayInfo, usage *dto.Usage, modelRatio, groupRatio, completionRatio, audioRatio, audioCompletionRatio, modelPrice float64) map[string]interface{} {
-	info := GenerateTextOtherInfo(ctx, relayInfo, modelRatio, groupRatio, completionRatio, modelPrice)
+	info := GenerateTextOtherInfo(ctx, relayInfo, modelRatio, groupRatio, completionRatio, 0, 0.0, modelPrice)
 	info["audio"] = true
 	info["audio_input"] = usage.PromptTokensDetails.AudioTokens
 	info["audio_output"] = usage.CompletionTokenDetails.AudioTokens

service/quota.go

@@ -12,6 +12,7 @@ import (
 	relaycommon "one-api/relay/common"
 	"one-api/relay/helper"
 	"one-api/setting"
+	"one-api/setting/operation_setting"
 	"strings"
 	"time"
 
@@ -38,9 +39,9 @@ func calculateAudioQuota(info QuotaInfo) int {
 		return int(info.ModelPrice * common.QuotaPerUnit * info.GroupRatio)
 	}
 
-	completionRatio := common.GetCompletionRatio(info.ModelName)
-	audioRatio := common.GetAudioRatio(info.ModelName)
-	audioCompletionRatio := common.GetAudioCompletionRatio(info.ModelName)
+	completionRatio := operation_setting.GetCompletionRatio(info.ModelName)
+	audioRatio := operation_setting.GetAudioRatio(info.ModelName)
+	audioCompletionRatio := operation_setting.GetAudioCompletionRatio(info.ModelName)
 	ratio := info.GroupRatio * info.ModelRatio
 
 	quota := info.InputDetails.TextTokens + int(math.Round(float64(info.OutputDetails.TextTokens)*completionRatio))
@@ -75,7 +76,7 @@ func PreWssConsumeQuota(ctx *gin.Context, relayInfo *relaycommon.RelayInfo, usag
 	audioInputTokens := usage.InputTokenDetails.AudioTokens
 	audioOutTokens := usage.OutputTokenDetails.AudioTokens
 	groupRatio := setting.GetGroupRatio(relayInfo.Group)
-	modelRatio := common.GetModelRatio(modelName)
+	modelRatio, _ := operation_setting.GetModelRatio(modelName)
 
 	quotaInfo := QuotaInfo{
 		InputDetails: TokenDetails{
@@ -122,9 +123,9 @@ func PostWssConsumeQuota(ctx *gin.Context, relayInfo *relaycommon.RelayInfo, mod
 	audioOutTokens := usage.OutputTokenDetails.AudioTokens
 
 	tokenName := ctx.GetString("token_name")
-	completionRatio := common.GetCompletionRatio(modelName)
-	audioRatio := common.GetAudioRatio(relayInfo.OriginModelName)
-	audioCompletionRatio := common.GetAudioCompletionRatio(modelName)
+	completionRatio := operation_setting.GetCompletionRatio(modelName)
+	audioRatio := operation_setting.GetAudioRatio(relayInfo.OriginModelName)
+	audioCompletionRatio := operation_setting.GetAudioCompletionRatio(modelName)
 
 	quotaInfo := QuotaInfo{
 		InputDetails: TokenDetails{
@@ -184,9 +185,9 @@ func PostAudioConsumeQuota(ctx *gin.Context, relayInfo *relaycommon.RelayInfo,
 	audioOutTokens := usage.CompletionTokenDetails.AudioTokens
 
 	tokenName := ctx.GetString("token_name")
-	completionRatio := common.GetCompletionRatio(relayInfo.OriginModelName)
-	audioRatio := common.GetAudioRatio(relayInfo.OriginModelName)
-	audioCompletionRatio := common.GetAudioCompletionRatio(relayInfo.OriginModelName)
+	completionRatio := operation_setting.GetCompletionRatio(relayInfo.OriginModelName)
+	audioRatio := operation_setting.GetAudioRatio(relayInfo.OriginModelName)
+	audioCompletionRatio := operation_setting.GetAudioCompletionRatio(relayInfo.OriginModelName)
 
 	modelRatio := priceData.ModelRatio
 	groupRatio := priceData.GroupRatio

service/token_counter.go

@@ -1,7 +1,6 @@
 package service
 
 import (
-	"encoding/json"
 	"errors"
 	"fmt"
 	"image"
@@ -11,6 +10,7 @@ import (
 	"one-api/constant"
 	"one-api/dto"
 	relaycommon "one-api/relay/common"
+	"one-api/setting/operation_setting"
 	"strings"
 	"unicode/utf8"
 
@@ -33,7 +33,7 @@ func InitTokenEncoders() {
 	if err != nil {
 		common.FatalLog(fmt.Sprintf("failed to get gpt-4o token encoder: %s", err.Error()))
 	}
-	for model, _ := range common.GetDefaultModelRatioMap() {
+	for model, _ := range operation_setting.GetDefaultModelRatioMap() {
 		if strings.HasPrefix(model, "gpt-3.5") {
 			tokenEncoderMap[model] = cl100TokenEncoder
 		} else if strings.HasPrefix(model, "gpt-4") {
@@ -170,12 +170,7 @@ func CountTokenChatRequest(info *relaycommon.RelayInfo, request dto.GeneralOpenA
 	}
 	tkm += msgTokens
 	if request.Tools != nil {
-		toolsData, _ := json.Marshal(request.Tools)
-		var openaiTools []dto.OpenAITools
-		err := json.Unmarshal(toolsData, &openaiTools)
-		if err != nil {
-			return 0, errors.New(fmt.Sprintf("count_tools_token_fail: %s", err.Error()))
-		}
+		openaiTools := request.Tools
 		countStr := ""
 		for _, tool := range openaiTools {
 			countStr = tool.Function.Name

service/user_notify.go

@@ -11,7 +11,10 @@ import (
 
 func NotifyRootUser(t string, subject string, content string) {
 	user := model.GetRootUser().ToBaseUser()
-	_ = NotifyUser(user.Id, user.Email, user.GetSetting(), dto.NewNotify(t, subject, content, nil))
+	err := NotifyUser(user.Id, user.Email, user.GetSetting(), dto.NewNotify(t, subject, content, nil))
+	if err != nil {
+		common.SysError(fmt.Sprintf("failed to notify root user: %s", err.Error()))
+	}
 }
 
 func NotifyUser(userId int, userEmail string, userSetting map[string]interface{}, data dto.Notify) error {

setting/config/config.go

@@ -0,0 +1,259 @@
+package config
+
+import (
+	"encoding/json"
+	"one-api/common"
+	"reflect"
+	"strconv"
+	"strings"
+	"sync"
+)
+
+// ConfigManager 统一管理所有配置
+type ConfigManager struct {
+	configs map[string]interface{}
+	mutex   sync.RWMutex
+}
+
+var GlobalConfig = NewConfigManager()
+
+func NewConfigManager() *ConfigManager {
+	return &ConfigManager{
+		configs: make(map[string]interface{}),
+	}
+}
+
+// Register 注册一个配置模块
+func (cm *ConfigManager) Register(name string, config interface{}) {
+	cm.mutex.Lock()
+	defer cm.mutex.Unlock()
+	cm.configs[name] = config
+}
+
+// Get 获取指定配置模块
+func (cm *ConfigManager) Get(name string) interface{} {
+	cm.mutex.RLock()
+	defer cm.mutex.RUnlock()
+	return cm.configs[name]
+}
+
+// LoadFromDB 从数据库加载配置
+func (cm *ConfigManager) LoadFromDB(options map[string]string) error {
+	cm.mutex.Lock()
+	defer cm.mutex.Unlock()
+
+	for name, config := range cm.configs {
+		prefix := name + "."
+		configMap := make(map[string]string)
+
+		// 收集属于此配置的所有选项
+		for key, value := range options {
+			if strings.HasPrefix(key, prefix) {
+				configKey := strings.TrimPrefix(key, prefix)
+				configMap[configKey] = value
+			}
+		}
+
+		// 如果找到配置项，则更新配置
+		if len(configMap) > 0 {
+			if err := updateConfigFromMap(config, configMap); err != nil {
+				common.SysError("failed to update config " + name + ": " + err.Error())
+				continue
+			}
+		}
+	}
+
+	return nil
+}
+
+// SaveToDB 将配置保存到数据库
+func (cm *ConfigManager) SaveToDB(updateFunc func(key, value string) error) error {
+	cm.mutex.RLock()
+	defer cm.mutex.RUnlock()
+
+	for name, config := range cm.configs {
+		configMap, err := configToMap(config)
+		if err != nil {
+			return err
+		}
+
+		for key, value := range configMap {
+			dbKey := name + "." + key
+			if err := updateFunc(dbKey, value); err != nil {
+				return err
+			}
+		}
+	}
+
+	return nil
+}
+
+// 辅助函数：将配置对象转换为map
+func configToMap(config interface{}) (map[string]string, error) {
+	result := make(map[string]string)
+
+	val := reflect.ValueOf(config)
+	if val.Kind() == reflect.Ptr {
+		val = val.Elem()
+	}
+
+	if val.Kind() != reflect.Struct {
+		return nil, nil
+	}
+
+	typ := val.Type()
+	for i := 0; i < val.NumField(); i++ {
+		field := val.Field(i)
+		fieldType := typ.Field(i)
+
+		// 跳过未导出字段
+		if !fieldType.IsExported() {
+			continue
+		}
+
+		// 获取json标签作为键名
+		key := fieldType.Tag.Get("json")
+		if key == "" || key == "-" {
+			key = fieldType.Name
+		}
+
+		// 处理不同类型的字段
+		var strValue string
+		switch field.Kind() {
+		case reflect.String:
+			strValue = field.String()
+		case reflect.Bool:
+			strValue = strconv.FormatBool(field.Bool())
+		case reflect.Int, reflect.Int8, reflect.Int16, reflect.Int32, reflect.Int64:
+			strValue = strconv.FormatInt(field.Int(), 10)
+		case reflect.Uint, reflect.Uint8, reflect.Uint16, reflect.Uint32, reflect.Uint64:
+			strValue = strconv.FormatUint(field.Uint(), 10)
+		case reflect.Float32, reflect.Float64:
+			strValue = strconv.FormatFloat(field.Float(), 'f', -1, 64)
+		case reflect.Map, reflect.Slice, reflect.Struct:
+			// 复杂类型使用JSON序列化
+			bytes, err := json.Marshal(field.Interface())
+			if err != nil {
+				return nil, err
+			}
+			strValue = string(bytes)
+		default:
+			// 跳过不支持的类型
+			continue
+		}
+
+		result[key] = strValue
+	}
+
+	return result, nil
+}
+
+// 辅助函数：从map更新配置对象
+func updateConfigFromMap(config interface{}, configMap map[string]string) error {
+	val := reflect.ValueOf(config)
+	if val.Kind() != reflect.Ptr {
+		return nil
+	}
+	val = val.Elem()
+
+	if val.Kind() != reflect.Struct {
+		return nil
+	}
+
+	typ := val.Type()
+	for i := 0; i < val.NumField(); i++ {
+		field := val.Field(i)
+		fieldType := typ.Field(i)
+
+		// 跳过未导出字段
+		if !fieldType.IsExported() {
+			continue
+		}
+
+		// 获取json标签作为键名
+		key := fieldType.Tag.Get("json")
+		if key == "" || key == "-" {
+			key = fieldType.Name
+		}
+
+		// 检查map中是否有对应的值
+		strValue, ok := configMap[key]
+		if !ok {
+			continue
+		}
+
+		// 根据字段类型设置值
+		if !field.CanSet() {
+			continue
+		}
+
+		switch field.Kind() {
+		case reflect.String:
+			field.SetString(strValue)
+		case reflect.Bool:
+			boolValue, err := strconv.ParseBool(strValue)
+			if err != nil {
+				continue
+			}
+			field.SetBool(boolValue)
+		case reflect.Int, reflect.Int8, reflect.Int16, reflect.Int32, reflect.Int64:
+			intValue, err := strconv.ParseInt(strValue, 10, 64)
+			if err != nil {
+				continue
+			}
+			field.SetInt(intValue)
+		case reflect.Uint, reflect.Uint8, reflect.Uint16, reflect.Uint32, reflect.Uint64:
+			uintValue, err := strconv.ParseUint(strValue, 10, 64)
+			if err != nil {
+				continue
+			}
+			field.SetUint(uintValue)
+		case reflect.Float32, reflect.Float64:
+			floatValue, err := strconv.ParseFloat(strValue, 64)
+			if err != nil {
+				continue
+			}
+			field.SetFloat(floatValue)
+		case reflect.Map, reflect.Slice, reflect.Struct:
+			// 复杂类型使用JSON反序列化
+			err := json.Unmarshal([]byte(strValue), field.Addr().Interface())
+			if err != nil {
+				continue
+			}
+		}
+	}
+
+	return nil
+}
+
+// ConfigToMap 将配置对象转换为map（导出函数）
+func ConfigToMap(config interface{}) (map[string]string, error) {
+	return configToMap(config)
+}
+
+// UpdateConfigFromMap 从map更新配置对象（导出函数）
+func UpdateConfigFromMap(config interface{}, configMap map[string]string) error {
+	return updateConfigFromMap(config, configMap)
+}
+
+// ExportAllConfigs 导出所有已注册的配置为扁平结构
+func (cm *ConfigManager) ExportAllConfigs() map[string]string {
+	cm.mutex.RLock()
+	defer cm.mutex.RUnlock()
+
+	result := make(map[string]string)
+
+	for name, cfg := range cm.configs {
+		configMap, err := ConfigToMap(cfg)
+		if err != nil {
+			continue
+		}
+
+		// 使用 "模块名.配置项" 的格式添加到结果中
+		for key, value := range configMap {
+			result[name+"."+key] = value
+		}
+	}
+
+	return result
+}

setting/model_setting.go

@@ -1,45 +0,0 @@
-package setting
-
-import (
-	"encoding/json"
-	"one-api/common"
-)
-
-var geminiSafetySettings = map[string]string{
-	"default":                       "OFF",
-	"HARM_CATEGORY_CIVIC_INTEGRITY": "BLOCK_NONE",
-}
-
-func GetGeminiSafetySetting(key string) string {
-	if value, ok := geminiSafetySettings[key]; ok {
-		return value
-	}
-	return geminiSafetySettings["default"]
-}
-
-func GeminiSafetySettingFromJsonString(jsonString string) {
-	geminiSafetySettings = map[string]string{}
-	err := json.Unmarshal([]byte(jsonString), &geminiSafetySettings)
-	if err != nil {
-		geminiSafetySettings = map[string]string{
-			"default":                       "OFF",
-			"HARM_CATEGORY_CIVIC_INTEGRITY": "BLOCK_NONE",
-		}
-	}
-	// check must have default
-	if _, ok := geminiSafetySettings["default"]; !ok {
-		geminiSafetySettings["default"] = common.GeminiSafetySetting
-	}
-}
-
-func GeminiSafetySettingsJsonString() string {
-	// check must have default
-	if _, ok := geminiSafetySettings["default"]; !ok {
-		geminiSafetySettings["default"] = common.GeminiSafetySetting
-	}
-	jsonString, err := json.Marshal(geminiSafetySettings)
-	if err != nil {
-		return "{}"
-	}
-	return string(jsonString)
-}

setting/model_setting/claude.go

@@ -0,0 +1,65 @@
+package model_setting
+
+import (
+	"net/http"
+	"one-api/setting/config"
+)
+
+//var claudeHeadersSettings = map[string][]string{}
+//
+//var ClaudeThinkingAdapterEnabled = true
+//var ClaudeThinkingAdapterMaxTokens = 8192
+//var ClaudeThinkingAdapterBudgetTokensPercentage = 0.8
+
+// ClaudeSettings 定义Claude模型的配置
+type ClaudeSettings struct {
+	HeadersSettings                       map[string]map[string][]string `json:"model_headers_settings"`
+	DefaultMaxTokens                      map[string]int                 `json:"default_max_tokens"`
+	ThinkingAdapterEnabled                bool                           `json:"thinking_adapter_enabled"`
+	ThinkingAdapterBudgetTokensPercentage float64                        `json:"thinking_adapter_budget_tokens_percentage"`
+}
+
+// 默认配置
+var defaultClaudeSettings = ClaudeSettings{
+	HeadersSettings:        map[string]map[string][]string{},
+	ThinkingAdapterEnabled: true,
+	DefaultMaxTokens: map[string]int{
+		"default": 8192,
+	},
+	ThinkingAdapterBudgetTokensPercentage: 0.8,
+}
+
+// 全局实例
+var claudeSettings = defaultClaudeSettings
+
+func init() {
+	// 注册到全局配置管理器
+	config.GlobalConfig.Register("claude", &claudeSettings)
+}
+
+// GetClaudeSettings 获取Claude配置
+func GetClaudeSettings() *ClaudeSettings {
+	// check default max tokens must have default key
+	if _, ok := claudeSettings.DefaultMaxTokens["default"]; !ok {
+		claudeSettings.DefaultMaxTokens["default"] = 8192
+	}
+	return &claudeSettings
+}
+
+func (c *ClaudeSettings) WriteHeaders(originModel string, httpHeader *http.Header) {
+	if headers, ok := c.HeadersSettings[originModel]; ok {
+		for headerKey, headerValues := range headers {
+			httpHeader.Del(headerKey)
+			for _, headerValue := range headerValues {
+				httpHeader.Add(headerKey, headerValue)
+			}
+		}
+	}
+}
+
+func (c *ClaudeSettings) GetDefaultMaxTokens(model string) int {
+	if maxTokens, ok := c.DefaultMaxTokens[model]; ok {
+		return maxTokens
+	}
+	return c.DefaultMaxTokens["default"]
+}

setting/model_setting/gemini.go

@@ -0,0 +1,52 @@
+package model_setting
+
+import (
+	"one-api/setting/config"
+)
+
+// GeminiSettings 定义Gemini模型的配置
+type GeminiSettings struct {
+	SafetySettings  map[string]string `json:"safety_settings"`
+	VersionSettings map[string]string `json:"version_settings"`
+}
+
+// 默认配置
+var defaultGeminiSettings = GeminiSettings{
+	SafetySettings: map[string]string{
+		"default":                       "OFF",
+		"HARM_CATEGORY_CIVIC_INTEGRITY": "BLOCK_NONE",
+	},
+	VersionSettings: map[string]string{
+		"default":        "v1beta",
+		"gemini-1.0-pro": "v1",
+	},
+}
+
+// 全局实例
+var geminiSettings = defaultGeminiSettings
+
+func init() {
+	// 注册到全局配置管理器
+	config.GlobalConfig.Register("gemini", &geminiSettings)
+}
+
+// GetGeminiSettings 获取Gemini配置
+func GetGeminiSettings() *GeminiSettings {
+	return &geminiSettings
+}
+
+// GetGeminiSafetySetting 获取安全设置
+func GetGeminiSafetySetting(key string) string {
+	if value, ok := geminiSettings.SafetySettings[key]; ok {
+		return value
+	}
+	return geminiSettings.SafetySettings["default"]
+}
+
+// GetGeminiVersionSetting 获取版本设置
+func GetGeminiVersionSetting(key string) string {
+	if value, ok := geminiSettings.VersionSettings[key]; ok {
+		return value
+	}
+	return geminiSettings.VersionSettings["default"]
+}

setting/operation_setting/cache_ratio.go

@@ -0,0 +1,77 @@
+package operation_setting
+
+import (
+	"encoding/json"
+	"one-api/common"
+	"sync"
+)
+
+var defaultCacheRatio = map[string]float64{
+	"gpt-4":                        0.5,
+	"o1-2024-12-17":                0.5,
+	"o1-preview-2024-09-12":        0.5,
+	"o1-mini-2024-09-12":           0.5,
+	"gpt-4o-2024-11-20":            0.5,
+	"gpt-4o-2024-08-06":            0.5,
+	"gpt-4o-mini-2024-07-18":       0.5,
+	"gpt-4o-realtime-preview":      0.5,
+	"gpt-4o-mini-realtime-preview": 0.5,
+	"deepseek-chat":                0.5,
+	"deepseek-reasoner":            0.5,
+	"deepseek-coder":               0.5,
+}
+
+var cacheRatioMap map[string]float64
+var cacheRatioMapMutex sync.RWMutex
+
+// GetCacheRatioMap returns the cache ratio map
+func GetCacheRatioMap() map[string]float64 {
+	cacheRatioMapMutex.Lock()
+	defer cacheRatioMapMutex.Unlock()
+	if cacheRatioMap == nil {
+		cacheRatioMap = defaultCacheRatio
+	}
+	return cacheRatioMap
+}
+
+// CacheRatio2JSONString converts the cache ratio map to a JSON string
+func CacheRatio2JSONString() string {
+	GetCacheRatioMap()
+	jsonBytes, err := json.Marshal(cacheRatioMap)
+	if err != nil {
+		common.SysError("error marshalling cache ratio: " + err.Error())
+	}
+	return string(jsonBytes)
+}
+
+// UpdateCacheRatioByJSONString updates the cache ratio map from a JSON string
+func UpdateCacheRatioByJSONString(jsonStr string) error {
+	cacheRatioMapMutex.Lock()
+	defer cacheRatioMapMutex.Unlock()
+	cacheRatioMap = make(map[string]float64)
+	return json.Unmarshal([]byte(jsonStr), &cacheRatioMap)
+}
+
+// GetCacheRatio returns the cache ratio for a model
+func GetCacheRatio(name string) (float64, bool) {
+	GetCacheRatioMap()
+	ratio, ok := cacheRatioMap[name]
+	if !ok {
+		return 0.5, false // Default to 0.5 if not found
+	}
+	return ratio, true
+}
+
+// DefaultCacheRatio2JSONString converts the default cache ratio map to a JSON string
+func DefaultCacheRatio2JSONString() string {
+	jsonBytes, err := json.Marshal(defaultCacheRatio)
+	if err != nil {
+		common.SysError("error marshalling default cache ratio: " + err.Error())
+	}
+	return string(jsonBytes)
+}
+
+// GetDefaultCacheRatioMap returns the default cache ratio map
+func GetDefaultCacheRatioMap() map[string]float64 {
+	return defaultCacheRatio
+}

setting/operation_setting/model-ratio.go

@@ -1,7 +1,8 @@
-package common
+package operation_setting
 
 import (
 	"encoding/json"
+	"one-api/common"
 	"strings"
 	"sync"
 )
@@ -50,24 +51,26 @@ var defaultModelRatio = map[string]float64{
 	"gpt-4o-realtime-preview-2024-12-17":      2.5,
 	"gpt-4o-mini-realtime-preview":            0.3,
 	"gpt-4o-mini-realtime-preview-2024-12-17": 0.3,
-	"o1":                        7.5,
-	"o1-2024-12-17":             7.5,
-	"o1-preview":                7.5,
-	"o1-preview-2024-09-12":     7.5,
-	"o1-mini":                   0.55,
-	"o1-mini-2024-09-12":        0.55,
-	"o3-mini":                   0.55,
-	"o3-mini-2025-01-31":        0.55,
-	"o3-mini-high":              0.55,
-	"o3-mini-2025-01-31-high":   0.55,
-	"o3-mini-low":               0.55,
-	"o3-mini-2025-01-31-low":    0.55,
-	"o3-mini-medium":            0.55,
-	"o3-mini-2025-01-31-medium": 0.55,
-	"gpt-4o-mini":               0.075,
-	"gpt-4o-mini-2024-07-18":    0.075,
-	"gpt-4-turbo":               5, // $0.01 / 1K tokens
-	"gpt-4-turbo-2024-04-09":    5, // $0.01 / 1K tokens
+	"o1":                         7.5,
+	"o1-2024-12-17":              7.5,
+	"o1-preview":                 7.5,
+	"o1-preview-2024-09-12":      7.5,
+	"o1-mini":                    0.55,
+	"o1-mini-2024-09-12":         0.55,
+	"o3-mini":                    0.55,
+	"o3-mini-2025-01-31":         0.55,
+	"o3-mini-high":               0.55,
+	"o3-mini-2025-01-31-high":    0.55,
+	"o3-mini-low":                0.55,
+	"o3-mini-2025-01-31-low":     0.55,
+	"o3-mini-medium":             0.55,
+	"o3-mini-2025-01-31-medium":  0.55,
+	"gpt-4o-mini":                0.075,
+	"gpt-4o-mini-2024-07-18":     0.075,
+	"gpt-4-turbo":                5, // $0.01 / 1K tokens
+	"gpt-4-turbo-2024-04-09":     5, // $0.01 / 1K tokens
+	"gpt-4.5-preview":            37.5,
+	"gpt-4.5-preview-2025-02-27": 37.5,
 	//"gpt-3.5-turbo-0301":           0.75, //deprecated
 	"gpt-3.5-turbo":          0.25,
 	"gpt-3.5-turbo-0613":     0.75,
@@ -259,7 +262,7 @@ func ModelPrice2JSONString() string {
 	GetModelPriceMap()
 	jsonBytes, err := json.Marshal(modelPriceMap)
 	if err != nil {
-		SysError("error marshalling model price: " + err.Error())
+		common.SysError("error marshalling model price: " + err.Error())
 	}
 	return string(jsonBytes)
 }
@@ -283,7 +286,7 @@ func GetModelPrice(name string, printErr bool) (float64, bool) {
 	price, ok := modelPriceMap[name]
 	if !ok {
 		if printErr {
-			SysError("model price not found: " + name)
+			common.SysError("model price not found: " + name)
 		}
 		return -1, false
 	}
@@ -303,7 +306,7 @@ func ModelRatio2JSONString() string {
 	GetModelRatioMap()
 	jsonBytes, err := json.Marshal(modelRatioMap)
 	if err != nil {
-		SysError("error marshalling model ratio: " + err.Error())
+		common.SysError("error marshalling model ratio: " + err.Error())
 	}
 	return string(jsonBytes)
 }
@@ -315,23 +318,22 @@ func UpdateModelRatioByJSONString(jsonStr string) error {
 	return json.Unmarshal([]byte(jsonStr), &modelRatioMap)
 }
 
-func GetModelRatio(name string) float64 {
+func GetModelRatio(name string) (float64, bool) {
 	GetModelRatioMap()
 	if strings.HasPrefix(name, "gpt-4-gizmo") {
 		name = "gpt-4-gizmo-*"
 	}
 	ratio, ok := modelRatioMap[name]
 	if !ok {
-		SysError("model ratio not found: " + name)
-		return 30
+		return 37.5, SelfUseModeEnabled
 	}
-	return ratio
+	return ratio, true
 }
 
 func DefaultModelRatio2JSONString() string {
 	jsonBytes, err := json.Marshal(defaultModelRatio)
 	if err != nil {
-		SysError("error marshalling model ratio: " + err.Error())
+		common.SysError("error marshalling model ratio: " + err.Error())
 	}
 	return string(jsonBytes)
 }
@@ -353,7 +355,7 @@ func CompletionRatio2JSONString() string {
 	GetCompletionRatioMap()
 	jsonBytes, err := json.Marshal(CompletionRatio)
 	if err != nil {
-		SysError("error marshalling completion ratio: " + err.Error())
+		common.SysError("error marshalling completion ratio: " + err.Error())
 	}
 	return string(jsonBytes)
 }
@@ -387,6 +389,9 @@ func GetCompletionRatio(name string) float64 {
 			}
 			return 4
 		}
+		if strings.HasPrefix(name, "gpt-4.5") {
+			return 2
+		}
 		if strings.HasPrefix(name, "gpt-4-turbo") || strings.HasSuffix(name, "preview") {
 			return 3
 		}

setting/operation_setting/operation_setting.go

@@ -1,8 +1,9 @@
-package setting
+package operation_setting
 
 import "strings"
 
 var DemoSiteEnabled = false
+var SelfUseModeEnabled = false
 
 var AutomaticDisableKeywords = []string{
 	"Your credit balance is too low",

web/src/App.js

@@ -30,6 +30,7 @@ import { useTranslation } from 'react-i18next';
 import { StatusContext } from './context/Status';
 import { setStatusData } from './helpers/data.js';
 import { API, showError } from './helpers';
+import PersonalSetting from './components/PersonalSetting.js';
 
 const Home = lazy(() => import('./pages/Home'));
 const Detail = lazy(() => import('./pages/Detail'));
@@ -177,6 +178,16 @@ function App() {
             </PrivateRoute>
           }
         />
+        <Route
+          path='/personal'
+          element={
+            <PrivateRoute>
+              <Suspense fallback={<Loading></Loading>}>
+                <PersonalSetting />
+              </Suspense>
+            </PrivateRoute>
+          }
+        />
         <Route
           path='/topup'
           element={

web/src/components/ChannelsTable.js

@@ -15,7 +15,7 @@ import {
   getQuotaPerUnit,
   renderGroup,
   renderNumberWithPoint,
-  renderQuota, renderQuotaWithPrompt
+  renderQuota, renderQuotaWithPrompt, stringToColor
 } from '../helpers/render';
 import {
   Button, Divider,
@@ -378,17 +378,15 @@ const ChannelsTable = () => {
                 >
                   {t('测试')}
                 </Button>
-                <Dropdown
-                  trigger="click"
-                  position="bottomRight"
-                  menu={modelMenuItems}  // 使用即时生成的菜单项
-                >
-                  <Button
-                    style={{ padding: '8px 4px' }}
-                    type="primary"
-                    icon={<IconTreeTriangleDown />}
-                  ></Button>
-                </Dropdown>
+                <Button
+                  style={{ padding: '8px 4px' }}
+                  type="primary"
+                  icon={<IconTreeTriangleDown />}
+                  onClick={() => {
+                    setCurrentTestChannel(record);
+                    setShowModelTestModal(true);
+                  }}
+                ></Button>
               </SplitButtonGroup>
               <Popconfirm
                 title={t('确定是否要删除此渠道？')}
@@ -522,6 +520,9 @@ const ChannelsTable = () => {
   const [enableTagMode, setEnableTagMode] = useState(false);
   const [showBatchSetTag, setShowBatchSetTag] = useState(false);
   const [batchSetTagValue, setBatchSetTagValue] = useState('');
+  const [showModelTestModal, setShowModelTestModal] = useState(false);
+  const [currentTestChannel, setCurrentTestChannel] = useState(null);
+  const [modelSearchKeyword, setModelSearchKeyword] = useState('');
 
 
   const removeRecord = (record) => {
@@ -1289,6 +1290,77 @@ const ChannelsTable = () => {
           onChange={(v) => setBatchSetTagValue(v)}
         />
       </Modal>
+      
+      {/* 模型测试弹窗 */}
+      <Modal
+        title={t('选择模型进行测试')}
+        visible={showModelTestModal && currentTestChannel !== null}
+        onCancel={() => {
+          setShowModelTestModal(false);
+          setModelSearchKeyword('');
+        }}
+        footer={null}
+        maskClosable={true}
+        centered={true}
+        width={600}
+      >
+        <div style={{ maxHeight: '500px', overflowY: 'auto', padding: '10px' }}>
+          {currentTestChannel && (
+            <div>
+              <Typography.Title heading={6} style={{ marginBottom: '16px' }}>
+                {t('渠道')}: {currentTestChannel.name}
+              </Typography.Title>
+              
+              {/* 搜索框 */}
+              <Input
+                placeholder={t('搜索模型...')}
+                value={modelSearchKeyword}
+                onChange={(value) => setModelSearchKeyword(value)}
+                style={{ marginBottom: '16px' }}
+                showClear
+              />
+              
+              <div style={{ 
+                display: 'grid', 
+                gridTemplateColumns: 'repeat(auto-fill, minmax(180px, 1fr))', 
+                gap: '10px' 
+              }}>
+                {currentTestChannel.models.split(',')
+                  .filter(model => model.toLowerCase().includes(modelSearchKeyword.toLowerCase()))
+                  .map((model, index) => {
+
+                    return (
+                      <Button
+                        key={index}
+                        theme="light"
+                        type="tertiary"
+                        style={{ 
+                          height: 'auto',
+                          padding: '8px 12px',
+                          textAlign: 'center',
+                        }}
+                        onClick={() => {
+                          testChannel(currentTestChannel, model);
+                        }}
+                      >
+                        {model}
+                      </Button>
+                    );
+                  })}
+              </div>
+              
+              {/* 显示搜索结果数量 */}
+              {modelSearchKeyword && (
+                <Typography.Text type="secondary" style={{ marginTop: '16px', display: 'block' }}>
+                  {t('找到')} {currentTestChannel.models.split(',').filter(model => 
+                    model.toLowerCase().includes(modelSearchKeyword.toLowerCase())
+                  ).length} {t('个模型')}
+                </Typography.Text>
+              )}
+            </div>
+          )}
+        </div>
+      </Modal>
     </>
   );
 };

web/src/components/HeaderBar.js

@@ -21,15 +21,17 @@ import {
   IconUser,
   IconLanguage
 } from '@douyinfe/semi-icons';
-import { Avatar, Button, Dropdown, Layout, Nav, Switch } from '@douyinfe/semi-ui';
+import { Avatar, Button, Dropdown, Layout, Nav, Switch, Tag } from '@douyinfe/semi-ui';
 import { stringToColor } from '../helpers/render';
 import Text from '@douyinfe/semi-ui/lib/es/typography/text';
 import { StyleContext } from '../context/Style/index.js';
+import { StatusContext } from '../context/Status/index.js';
 
 const HeaderBar = () => {
   const { t, i18n } = useTranslation();
   const [userState, userDispatch] = useContext(UserContext);
   const [styleState, styleDispatch] = useContext(StyleContext);
+  const [statusState, statusDispatch] = useContext(StatusContext);
   let navigate = useNavigate();
   const [currentLang, setCurrentLang] = useState(i18n.language);
 
@@ -40,6 +42,10 @@ const HeaderBar = () => {
   const isNewYear =
     (currentDate.getMonth() === 0 && currentDate.getDate() === 1);
 
+  // Check if self-use mode is enabled
+  const isSelfUseMode = statusState?.status?.self_use_mode_enabled || false;
+  const isDemoSiteMode = statusState?.status?.demo_site_enabled || false;
+
   let buttons = [
     {
       text: t('首页'),
@@ -166,7 +172,7 @@ const HeaderBar = () => {
             onSelect={(key) => {}}
             header={styleState.isMobile?{
               logo: (
-                <>
+                <div style={{ display: 'flex', alignItems: 'center', position: 'relative' }}>
                   {
                     !styleState.showSider ?
                       <Button icon={<IconMenu />} theme="light" aria-label={t('展开侧边栏')} onClick={
@@ -176,13 +182,52 @@ const HeaderBar = () => {
                         () => styleDispatch({ type: 'SET_SIDER', payload: false })
                       } />
                   }
-                </>
+                  {(isSelfUseMode || isDemoSiteMode) && (
+                    <Tag 
+                      color={isSelfUseMode ? 'purple' : 'blue'}
+                      style={{ 
+                        position: 'absolute',
+                        top: '-8px',
+                        right: '-15px',
+                        fontSize: '0.7rem',
+                        padding: '0 4px',
+                        height: 'auto',
+                        lineHeight: '1.2',
+                        zIndex: 1,
+                        pointerEvents: 'none'
+                      }}
+                    >
+                      {isSelfUseMode ? t('自用模式') : t('演示站点')}
+                    </Tag>
+                  )}
+                </div>
               ),
             }:{
               logo: (
                 <img src={logo} alt='logo' />
               ),
-              text: systemName,
+              text: (
+                <div style={{ position: 'relative', display: 'inline-block' }}>
+                  {systemName}
+                  {(isSelfUseMode || isDemoSiteMode) && (
+                    <Tag 
+                      color={isSelfUseMode ? 'purple' : 'blue'}
+                      style={{ 
+                        position: 'absolute', 
+                        top: '-10px', 
+                        right: '-25px', 
+                        fontSize: '0.7rem',
+                        padding: '0 4px',
+                        whiteSpace: 'nowrap',
+                        zIndex: 1,
+                        boxShadow: '0 0 3px rgba(255, 255, 255, 0.7)'
+                      }}
+                    >
+                      {isSelfUseMode ? t('自用模式') : t('演示站点')}
+                    </Tag>
+                  )}
+                </div>
+              ),
             }}
             items={buttons}
             footer={
@@ -266,7 +311,8 @@ const HeaderBar = () => {
                       icon={<IconUser />}
                     />
                     {
-                      !styleState.isMobile && (
+                      // Hide register option in self-use mode
+                      !styleState.isMobile && !isSelfUseMode && (
                         <Nav.Item
                           itemKey={'register'}
                           text={t('注册')}

web/src/components/LogsTable.js

@@ -464,6 +464,8 @@ const LogsTable = () => {
           other.model_ratio,
           other.model_price,
           other.group_ratio,
+          other.cache_tokens || 0,
+          other.cache_ratio || 1.0,
         );
         return (
             <Paragraph
@@ -665,6 +667,8 @@ const LogsTable = () => {
             other?.audio_ratio,
             other?.audio_completion_ratio,
             other.group_ratio,
+            other.cache_tokens || 0,
+            other.cache_ratio || 1.0,
           );
         } else {
           content = renderModelPrice(
@@ -674,6 +678,8 @@ const LogsTable = () => {
             other.model_price,
             other.completion_ratio,
             other.group_ratio,
+            other.cache_tokens || 0,
+            other.cache_ratio || 1.0,
           );
         }
         expandDataLocal.push({

web/src/components/ModelSetting.js

@@ -1,27 +1,21 @@
 import React, { useEffect, useState } from 'react';
 import { Card, Spin, Tabs } from '@douyinfe/semi-ui';
-import SettingsGeneral from '../pages/Setting/Operation/SettingsGeneral.js';
-import SettingsDrawing from '../pages/Setting/Operation/SettingsDrawing.js';
-import SettingsSensitiveWords from '../pages/Setting/Operation/SettingsSensitiveWords.js';
-import SettingsLog from '../pages/Setting/Operation/SettingsLog.js';
-import SettingsDataDashboard from '../pages/Setting/Operation/SettingsDataDashboard.js';
-import SettingsMonitoring from '../pages/Setting/Operation/SettingsMonitoring.js';
-import SettingsCreditLimit from '../pages/Setting/Operation/SettingsCreditLimit.js';
-import SettingsMagnification from '../pages/Setting/Operation/SettingsMagnification.js';
-import ModelSettingsVisualEditor from '../pages/Setting/Operation/ModelSettingsVisualEditor.js';
-import GroupRatioSettings from '../pages/Setting/Operation/GroupRatioSettings.js';
-import ModelRatioSettings from '../pages/Setting/Operation/ModelRatioSettings.js';
 
 
 import { API, showError, showSuccess } from '../helpers';
-import SettingsChats from '../pages/Setting/Operation/SettingsChats.js';
 import { useTranslation } from 'react-i18next';
 import SettingGeminiModel from '../pages/Setting/Model/SettingGeminiModel.js';
+import SettingClaudeModel from '../pages/Setting/Model/SettingClaudeModel.js';
 
 const ModelSetting = () => {
   const { t } = useTranslation();
   let [inputs, setInputs] = useState({
-    GeminiSafetySettings: '',
+    'gemini.safety_settings': '',
+    'gemini.version_settings': '',
+    'claude.model_headers_settings': '',
+    'claude.thinking_adapter_enabled': true,
+    'claude.default_max_tokens': '',
+    'claude.thinking_adapter_budget_tokens_percentage': 0.8,
   });
 
   let [loading, setLoading] = useState(false);
@@ -33,7 +27,10 @@ const ModelSetting = () => {
       let newInputs = {};
       data.forEach((item) => {
         if (
-          item.key === 'GeminiSafetySettings'
+          item.key === 'gemini.safety_settings' ||
+          item.key === 'gemini.version_settings' ||
+          item.key === 'claude.model_headers_settings'||
+          item.key === 'claude.default_max_tokens'
         ) {
           item.value = JSON.stringify(JSON.parse(item.value), null, 2);
         }
@@ -74,6 +71,10 @@ const ModelSetting = () => {
         <Card style={{ marginTop: '10px' }}>
           <SettingGeminiModel options={inputs} refresh={onRefresh} />
         </Card>
+        {/* Claude */}
+        <Card style={{ marginTop: '10px' }}>
+          <SettingClaudeModel options={inputs} refresh={onRefresh} />
+        </Card>
       </Spin>
     </>
   );

web/src/components/OperationSetting.js

@@ -16,6 +16,7 @@ import ModelRatioSettings from '../pages/Setting/Operation/ModelRatioSettings.js
 import { API, showError, showSuccess } from '../helpers';
 import SettingsChats from '../pages/Setting/Operation/SettingsChats.js';
 import { useTranslation } from 'react-i18next';
+import ModelRatioNotSetEditor from '../pages/Setting/Operation/ModelRationNotSetEditor.js';
 
 const OperationSetting = () => {
   const { t } = useTranslation();
@@ -27,6 +28,7 @@ const OperationSetting = () => {
     PreConsumedQuota: 0,
     StreamCacheQueueLength: 0,
     ModelRatio: '',
+    CacheRatio: '',
     CompletionRatio: '',
     ModelPrice: '',
     GroupRatio: '',
@@ -59,6 +61,7 @@ const OperationSetting = () => {
     RetryTimes: 0,
     Chats: "[]",
     DemoSiteEnabled: false,
+    SelfUseModeEnabled: false,
     AutomaticDisableKeywords: '',
   });
 
@@ -75,7 +78,8 @@ const OperationSetting = () => {
           item.key === 'GroupRatio' ||
           item.key === 'UserUsableGroups' ||
           item.key === 'CompletionRatio' ||
-          item.key === 'ModelPrice'
+          item.key === 'ModelPrice' ||
+          item.key === 'CacheRatio'
         ) {
           item.value = JSON.stringify(JSON.parse(item.value), null, 2);
         }
@@ -158,6 +162,9 @@ const OperationSetting = () => {
             <Tabs.TabPane tab={t('可视化倍率设置')} itemKey="visual">
               <ModelSettingsVisualEditor options={inputs} refresh={onRefresh} />
             </Tabs.TabPane>
+            <Tabs.TabPane tab={t('未设置倍率模型')} itemKey="unset_models">
+              <ModelRatioNotSetEditor options={inputs} refresh={onRefresh} />
+            </Tabs.TabPane>
           </Tabs>
         </Card>
       </Spin>

web/src/components/OtherSetting.js

@@ -1,8 +1,10 @@
-import React, { useEffect, useRef, useState } from 'react';
-import { Banner, Button, Col, Form, Row } from '@douyinfe/semi-ui';
-import { API, showError, showSuccess } from '../helpers';
+import React, { useContext, useEffect, useRef, useState } from 'react';
+import { Banner, Button, Col, Form, Row, Modal, Space } from '@douyinfe/semi-ui';
+import { API, showError, showSuccess, timestamp2string } from '../helpers';
 import { marked } from 'marked';
 import { useTranslation } from 'react-i18next';
+import { StatusContext } from '../context/Status/index.js';
+import Text from '@douyinfe/semi-ui/lib/es/typography/text';
 
 const OtherSetting = () => {
   const { t } = useTranslation();
@@ -16,6 +18,7 @@ const OtherSetting = () => {
   });
   let [loading, setLoading] = useState(false);
   const [showUpdateModal, setShowUpdateModal] = useState(false);
+  const [statusState, statusDispatch] = useContext(StatusContext);
   const [updateData, setUpdateData] = useState({
     tag_name: '',
     content: '',
@@ -43,6 +46,7 @@ const OtherSetting = () => {
     HomePageContent: false,
     About: false,
     Footer: false,
+    CheckUpdate: false
   });
   const handleInputChange = async (value, e) => {
     const name = e.target.id;
@@ -145,23 +149,48 @@ const OtherSetting = () => {
     }
   };
 
-  const openGitHubRelease = () => {
-    window.location = 'https://github.com/songquanpeng/one-api/releases/latest';
-  };
-
   const checkUpdate = async () => {
-    const res = await API.get(
-      'https://api.github.com/repos/songquanpeng/one-api/releases/latest',
-    );
-    const { tag_name, body } = res.data;
-    if (tag_name === process.env.REACT_APP_VERSION) {
-      showSuccess(`已是最新版本：${tag_name}`);
-    } else {
-      setUpdateData({
-        tag_name: tag_name,
-        content: marked.parse(body),
-      });
-      setShowUpdateModal(true);
+    try {
+      setLoadingInput((loadingInput) => ({ ...loadingInput, CheckUpdate: true }));
+      // Use a CORS proxy to avoid direct cross-origin requests to GitHub API
+      // Option 1: Use a public CORS proxy service
+      // const proxyUrl = 'https://cors-anywhere.herokuapp.com/';
+      // const res = await API.get(
+      //   `${proxyUrl}https://api.github.com/repos/Calcium-Ion/new-api/releases/latest`,
+      // );
+      
+      // Option 2: Use the JSON proxy approach which often works better with GitHub API
+      const res = await fetch(
+        'https://api.github.com/repos/Calcium-Ion/new-api/releases/latest',
+        {
+          headers: {
+            'Accept': 'application/json',
+            'Content-Type': 'application/json',
+            // Adding User-Agent which is often required by GitHub API
+            'User-Agent': 'new-api-update-checker'
+          }
+        }
+      ).then(response => response.json());
+      
+      // Option 3: Use a local proxy endpoint
+      // Create a cached version of the response to avoid frequent GitHub API calls
+      // const res = await API.get('/api/status/github-latest-release');
+
+      const { tag_name, body } = res;
+      if (tag_name === statusState?.status?.version) {
+        showSuccess(`已是最新版本：${tag_name}`);
+      } else {
+        setUpdateData({
+          tag_name: tag_name,
+          content: marked.parse(body),
+        });
+        setShowUpdateModal(true);
+      }
+    } catch (error) {
+      console.error('Failed to check for updates:', error);
+      showError('检查更新失败，请稍后再试');
+    } finally {
+      setLoadingInput((loadingInput) => ({ ...loadingInput, CheckUpdate: false }));
     }
   };
   const getOptions = async () => {
@@ -186,9 +215,41 @@ const OtherSetting = () => {
     getOptions();
   }, []);
 
+  // Function to open GitHub release page
+  const openGitHubRelease = () => {
+    window.open(`https://github.com/Calcium-Ion/new-api/releases/tag/${updateData.tag_name}`, '_blank');
+  };
+
+  const getStartTimeString = () => {
+    const timestamp = statusState?.status?.start_time;
+    return statusState.status ? timestamp2string(timestamp) : '';
+  };
+
   return (
     <Row>
       <Col span={24}>
+        {/* 版本信息 */}
+        <Form style={{ marginBottom: 15 }}>
+          <Form.Section text={t('系统信息')}>
+            <Row>
+              <Col span={16}>
+                <Space>
+                  <Text>
+                    {t('当前版本')}：{statusState?.status?.version || t('未知')}
+                  </Text>
+                  <Button type="primary" onClick={checkUpdate} loading={loadingInput['CheckUpdate']}>
+                    {t('检查更新')}
+                  </Button>
+                </Space>
+              </Col>
+            </Row>
+            <Row>
+              <Col span={16}>
+                <Text>{t('启动时间')}：{getStartTimeString()}</Text>
+              </Col>
+            </Row>
+          </Form.Section>
+        </Form>
         {/* 通用设置 */}
         <Form
           values={inputs}
@@ -282,28 +343,25 @@ const OtherSetting = () => {
           </Form.Section>
         </Form>
       </Col>
-      {/*<Modal*/}
-      {/*  onClose={() => setShowUpdateModal(false)}*/}
-      {/*  onOpen={() => setShowUpdateModal(true)}*/}
-      {/*  open={showUpdateModal}*/}
-      {/*>*/}
-      {/*  <Modal.Header>新版本：{updateData.tag_name}</Modal.Header>*/}
-      {/*  <Modal.Content>*/}
-      {/*    <Modal.Description>*/}
-      {/*      <div dangerouslySetInnerHTML={{ __html: updateData.content }}></div>*/}
-      {/*    </Modal.Description>*/}
-      {/*  </Modal.Content>*/}
-      {/*  <Modal.Actions>*/}
-      {/*    <Button onClick={() => setShowUpdateModal(false)}>关闭</Button>*/}
-      {/*    <Button*/}
-      {/*      content='详情'*/}
-      {/*      onClick={() => {*/}
-      {/*        setShowUpdateModal(false);*/}
-      {/*        openGitHubRelease();*/}
-      {/*      }}*/}
-      {/*    />*/}
-      {/*  </Modal.Actions>*/}
-      {/*</Modal>*/}
+      <Modal
+        title={t('新版本') + '：' + updateData.tag_name}
+        visible={showUpdateModal}
+        onCancel={() => setShowUpdateModal(false)}
+        footer={[
+          <Button 
+            key="details" 
+            type="primary" 
+            onClick={() => {
+              setShowUpdateModal(false);
+              openGitHubRelease();
+            }}
+          >
+            {t('详情')}
+          </Button>
+        ]}
+      >
+        <div dangerouslySetInnerHTML={{ __html: updateData.content }}></div>
+      </Modal>
     </Row>
   );
 };

web/src/components/PersonalSetting.js

@@ -69,7 +69,11 @@ const PersonalSetting = () => {
     const [models, setModels] = useState([]);
     const [openTransfer, setOpenTransfer] = useState(false);
     const [transferAmount, setTransferAmount] = useState(0);
-    const [isModelsExpanded, setIsModelsExpanded] = useState(false);
+    const [isModelsExpanded, setIsModelsExpanded] = useState(() => {
+        // Initialize from localStorage if available
+        const savedState = localStorage.getItem('modelsExpanded');
+        return savedState ? JSON.parse(savedState) : false;
+    });
     const MODELS_DISPLAY_COUNT = 10;  // 默认显示的模型数量
     const [notificationSettings, setNotificationSettings] = useState({
         warningType: 'email',
@@ -124,6 +128,11 @@ const PersonalSetting = () => {
         }
     }, [userState?.user?.setting]);
 
+    // Save models expanded state to localStorage whenever it changes
+    useEffect(() => {
+        localStorage.setItem('modelsExpanded', JSON.stringify(isModelsExpanded));
+    }, [isModelsExpanded]);
+
     const handleInputChange = (name, value) => {
         setInputs((inputs) => ({...inputs, [name]: value}));
     };
@@ -384,7 +393,7 @@ const PersonalSetting = () => {
                             </div>
                         </div>
                     </Modal>
-                    <div style={{marginTop: 20}}>
+                    <div>
                         <Card
                             title={
                                 <Card.Meta

web/src/components/SiderBar.js

@@ -28,7 +28,7 @@ import {
   IconSetting,
   IconUser
 } from '@douyinfe/semi-icons';
-import { Avatar, Dropdown, Layout, Nav, Switch } from '@douyinfe/semi-ui';
+import { Avatar, Dropdown, Layout, Nav, Switch, Divider } from '@douyinfe/semi-ui';
 import { setStatusData } from '../helpers/data.js';
 import { stringToColor } from '../helpers/render.js';
 import { useSetTheme, useTheme } from '../context/Theme/index.js';
@@ -65,35 +65,11 @@ const SiderBar = () => {
     pricing: '/pricing',
     task: '/task',
     playground: '/playground',
+    personal: '/personal',
   };
 
-  const headerButtons = useMemo(
+  const workspaceItems = useMemo(
     () => [
-      {
-        text: 'Playground',
-        itemKey: 'playground',
-        to: '/playground',
-        icon: <IconCommentStroked />,
-      },
-      {
-        text: t('渠道'),
-        itemKey: 'channel',
-        to: '/channel',
-        icon: <IconLayers />,
-        className: isAdmin() ? '' : 'tableHiddle',
-      },
-      {
-        text: t('聊天'),
-        itemKey: 'chat',
-        items: chatItems,
-        icon: <IconComment />,
-      },
-      {
-        text: t('令牌'),
-        itemKey: 'token',
-        to: '/token',
-        icon: <IconKey />,
-      },
       {
         text: t('数据看板'),
         itemKey: 'detail',
@@ -105,33 +81,19 @@ const SiderBar = () => {
             : 'tableHiddle',
       },
       {
-        text: t('兑换码'),
-        itemKey: 'redemption',
-        to: '/redemption',
-        icon: <IconGift />,
-        className: isAdmin() ? '' : 'tableHiddle',
+        text: t('API令牌'),
+        itemKey: 'token',
+        to: '/token',
+        icon: <IconKey />,
       },
       {
-        text: t('钱包'),
-        itemKey: 'topup',
-        to: '/topup',
-        icon: <IconCreditCard />,
-      },
-      {
-        text: t('用户管理'),
-        itemKey: 'user',
-        to: '/user',
-        icon: <IconUser />,
-        className: isAdmin() ? '' : 'tableHiddle',
-      },
-      {
-        text: t('日志'),
+        text: t('使用日志'),
         itemKey: 'log',
         to: '/log',
         icon: <IconHistogram />,
       },
       {
-        text: t('绘图'),
+        text: t('绘图日志'),
         itemKey: 'midjourney',
         to: '/midjourney',
         icon: <IconImage />,
@@ -141,75 +103,148 @@ const SiderBar = () => {
             : 'tableHiddle',
       },
       {
-        text: t('异步任务'),
+        text: t('任务日志'),
         itemKey: 'task',
         to: '/task',
         icon: <IconChecklistStroked />,
         className:
-            localStorage.getItem('enable_task') === 'true'
-                ? ''
-                : 'tableHiddle',
-      },
-      {
-        text: t('设置'),
-        itemKey: 'setting',
-        to: '/setting',
-        icon: <IconSetting />,
-      },
+          localStorage.getItem('enable_task') === 'true'
+            ? ''
+            : 'tableHiddle',
+      }
     ],
     [
       localStorage.getItem('enable_data_export'),
       localStorage.getItem('enable_drawing'),
       localStorage.getItem('enable_task'),
-      localStorage.getItem('chat_link'),
-      chatItems,
-      isAdmin(),
       t,
     ],
   );
 
+  const financeItems = useMemo(
+    () => [
+      {
+        text: t('钱包'),
+        itemKey: 'topup',
+        to: '/topup',
+        icon: <IconCreditCard />,
+      },
+      {
+        text: t('个人设置'),
+        itemKey: 'personal',
+        to: '/personal',
+        icon: <IconUser />,
+      },
+    ],
+    [t],
+  );
+
+  const adminItems = useMemo(
+    () => [
+      {
+        text: t('渠道'),
+        itemKey: 'channel',
+        to: '/channel',
+        icon: <IconLayers />,
+        className: isAdmin() ? '' : 'tableHiddle',
+      },
+      {
+        text: t('兑换码'),
+        itemKey: 'redemption',
+        to: '/redemption',
+        icon: <IconGift />,
+        className: isAdmin() ? '' : 'tableHiddle',
+      },
+      {
+        text: t('用户管理'),
+        itemKey: 'user',
+        to: '/user',
+        icon: <IconUser />,
+      },
+      {
+        text: t('系统设置'),
+        itemKey: 'setting',
+        to: '/setting',
+        icon: <IconSetting />,
+      },
+    ],
+    [isAdmin(), t],
+  );
+
+  const chatMenuItems = useMemo(
+    () => [
+      {
+        text: 'Playground',
+        itemKey: 'playground',
+        to: '/playground',
+        icon: <IconCommentStroked />,
+      },
+      {
+        text: t('聊天'),
+        itemKey: 'chat',
+        items: chatItems,
+        icon: <IconComment />,
+      },
+    ],
+    [chatItems, t],
+  );
+
   useEffect(() => {
     let localKey = window.location.pathname.split('/')[1];
     if (localKey === '') {
       localKey = 'home';
     }
     setSelectedKeys([localKey]);
-    
+
     let chatLink = localStorage.getItem('chat_link');
     if (!chatLink) {
-        let chats = localStorage.getItem('chats');
-        if (chats) {
-            // console.log(chats);
-            try {
-                chats = JSON.parse(chats);
-                if (Array.isArray(chats)) {
-                    let chatItems = [];
-                    for (let i = 0; i < chats.length; i++) {
-                        let chat = {};
-                        for (let key in chats[i]) {
-                            chat.text = key;
-                            chat.itemKey = 'chat' + i;
-                            chat.to = '/chat/' + i;
-                        }
-                        // setRouterMap({ ...routerMap, chat: '/chat/' + i })
-                        chatItems.push(chat);
-                    }
-                    setChatItems(chatItems);
-                }
-            } catch (e) {
-                console.error(e);
-                showError('聊天数据解析失败')
+      let chats = localStorage.getItem('chats');
+      if (chats) {
+        // console.log(chats);
+        try {
+          chats = JSON.parse(chats);
+          if (Array.isArray(chats)) {
+            let chatItems = [];
+            for (let i = 0; i < chats.length; i++) {
+              let chat = {};
+              for (let key in chats[i]) {
+                chat.text = key;
+                chat.itemKey = 'chat' + i;
+                chat.to = '/chat/' + i;
+              }
+              // setRouterMap({ ...routerMap, chat: '/chat/' + i })
+              chatItems.push(chat);
             }
+            setChatItems(chatItems);
+          }
+        } catch (e) {
+          console.error(e);
+          showError('聊天数据解析失败')
         }
+      }
     }
-    
+
     setIsCollapsed(localStorage.getItem('default_collapse_sidebar') === 'true');
   }, []);
 
+  // Custom divider style
+  const dividerStyle = {
+    margin: '8px 0',
+    opacity: 0.6,
+  };
+
+  // Custom group label style
+  const groupLabelStyle = {
+    padding: '8px 16px',
+    color: 'var(--semi-color-text-2)',
+    fontSize: '12px',
+    fontWeight: 'normal',
+  };
+
   return (
     <>
       <Nav
-        style={{ maxWidth: 220, height: '100%' }}
+        style={{ maxWidth: 200, height: '100%' }}
         defaultIsCollapsed={
           localStorage.getItem('default_collapse_sidebar') === 'true'
         }
@@ -219,27 +254,27 @@ const SiderBar = () => {
         }}
         selectedKeys={selectedKeys}
         renderWrapper={({ itemElement, isSubNav, isInSubNav, props }) => {
-            let chatLink = localStorage.getItem('chat_link');
-            if (!chatLink) {
-                let chats = localStorage.getItem('chats');
-                if (chats) {
-                    chats = JSON.parse(chats);
-                    if (Array.isArray(chats) && chats.length > 0) {
-                        for (let i = 0; i < chats.length; i++) {
-                            routerMap['chat' + i] = '/chat/' + i;
-                        }
-                        if (chats.length > 1) {
-                            // delete /chat
-                            if (routerMap['chat']) {
-                                delete routerMap['chat'];
-                            }
-                        } else {
-                            // rename /chat to /chat/0
-                            routerMap['chat'] = '/chat/0';
-                        }
-                    }
+          let chatLink = localStorage.getItem('chat_link');
+          if (!chatLink) {
+            let chats = localStorage.getItem('chats');
+            if (chats) {
+              chats = JSON.parse(chats);
+              if (Array.isArray(chats) && chats.length > 0) {
+                for (let i = 0; i < chats.length; i++) {
+                  routerMap['chat' + i] = '/chat/' + i;
                 }
+                if (chats.length > 1) {
+                  // delete /chat
+                  if (routerMap['chat']) {
+                    delete routerMap['chat'];
+                  }
+                } else {
+                  // rename /chat to /chat/0
+                  routerMap['chat'] = '/chat/0';
+                }
+              }
             }
+          }
           return (
             <Link
               style={{ textDecoration: 'none' }}
@@ -249,7 +284,6 @@ const SiderBar = () => {
             </Link>
           );
         }}
-        items={headerButtons}
         onSelect={(key) => {
           if (key.itemKey.toString().startsWith('chat')) {
             styleDispatch({ type: 'SET_INNER_PADDING', payload: false });
@@ -258,12 +292,101 @@ const SiderBar = () => {
           }
           setSelectedKeys([key.itemKey]);
         }}
-        footer={
-          <>
-          </>
-        }
       >
-        <Nav.Footer collapseButton={true}></Nav.Footer>
+        {/* Chat Section - Only show if there are chat items */}
+        {chatItems.length > 0 && (
+          <>
+            {chatMenuItems.map((item) => {
+              if (item.items && item.items.length > 0) {
+                return (
+                  <Nav.Sub
+                    key={item.itemKey}
+                    itemKey={item.itemKey}
+                    text={item.text}
+                    icon={item.icon}
+                  >
+                    {item.items.map((subItem) => (
+                      <Nav.Item
+                        key={subItem.itemKey}
+                        itemKey={subItem.itemKey}
+                        text={subItem.text}
+                      />
+                    ))}
+                  </Nav.Sub>
+                );
+              } else {
+                return (
+                  <Nav.Item
+                    key={item.itemKey}
+                    itemKey={item.itemKey}
+                    text={item.text}
+                    icon={item.icon}
+                  />
+                );
+              }
+            })}
+          </>
+        )}
+
+        {/* Divider */}
+        <Divider style={dividerStyle} />
+
+        {/* Workspace Section */}
+        {!isCollapsed && <div style={groupLabelStyle}>{t('控制台')}</div>}
+        {workspaceItems.map((item) => (
+          <Nav.Item
+            key={item.itemKey}
+            itemKey={item.itemKey}
+            text={item.text}
+            icon={item.icon}
+            className={item.className}
+          />
+        ))}
+
+        {/* Divider */}
+        <Divider style={dividerStyle} />
+
+        {/* Finance Management Section */}
+        {!isCollapsed && <div style={groupLabelStyle}>{t('个人中心')}</div>}
+        {financeItems.map((item) => (
+          <Nav.Item
+            key={item.itemKey}
+            itemKey={item.itemKey}
+            text={item.text}
+            icon={item.icon}
+            className={item.className}
+          />
+        ))}
+
+        {isAdmin() && (
+          <>
+            {/* Divider */}
+            <Divider style={dividerStyle} />
+
+            {/* Admin Section */}
+            {adminItems.map((item) => (
+              <Nav.Item
+                key={item.itemKey}
+                itemKey={item.itemKey}
+                text={item.text}
+                icon={item.icon}
+                className={item.className}
+              />
+            ))}
+          </>
+        )}
+
+        <Nav.Footer
+          collapseButton={true}
+          collapseText={(collapsed)=>
+            {
+              if(collapsed){
+                return t('展开侧边栏')
+              }
+                return t('收起侧边栏')
+            }
+          }
+        />
       </Nav>
     </>
   );

web/src/components/UsersTable.js

@@ -376,7 +376,7 @@ const UsersTable = () => {
     if (searchKeyword === '') {
       await loadUsers(activePage, pageSize);
     } else {
-      await searchUsers(searchKeyword, searchGroup);
+      await searchUsers(activePage, pageSize, searchKeyword, searchGroup);
     }
   };
 

web/src/constants/channel.constants.js

@@ -82,7 +82,7 @@ export const CHANNEL_OPTIONS = [
   {
     value: 45,
     color: 'blue',
-    label: '火山方舟（豆包）'
+    label: '字节火山方舟、豆包、DeepSeek通用'
   },
   { value: 25, color: 'green', label: 'Moonshot' },
   { value: 19, color: 'blue', label: '360 智脑' },

web/src/helpers/render.js

@@ -298,6 +298,8 @@ export function renderModelPrice(
   modelPrice = -1,
   completionRatio,
   groupRatio,
+  cacheTokens = 0,
+  cacheRatio = 1.0,
 ) {
   if (modelPrice !== -1) {
     return i18next.t('模型价格：${{price}} * 分组倍率：{{ratio}} = ${{total}}', {
@@ -311,9 +313,15 @@ export function renderModelPrice(
     }
     let inputRatioPrice = modelRatio * 2.0;
     let completionRatioPrice = modelRatio * 2.0 * completionRatio;
+    let cacheRatioPrice = modelRatio * 2.0 * cacheRatio;
+    
+    // Calculate effective input tokens (non-cached + cached with ratio applied)
+    const effectiveInputTokens = (inputTokens - cacheTokens) + (cacheTokens * cacheRatio);
+    
     let price =
-      (inputTokens / 1000000) * inputRatioPrice * groupRatio +
+      (effectiveInputTokens / 1000000) * inputRatioPrice * groupRatio +
       (completionTokens / 1000000) * completionRatioPrice * groupRatio;
+    
     return (
       <>
         <article>
@@ -327,16 +335,36 @@ export function renderModelPrice(
             ratio: groupRatio,
             total: completionRatioPrice * groupRatio
           })}</p>
+          {cacheTokens > 0 && (
+            <p>{i18next.t('缓存：${{price}} * {{ratio}} = ${{total}} / 1M tokens (缓存比例: {{cacheRatio}})', {
+              price: cacheRatioPrice,
+              ratio: groupRatio,
+              total: cacheRatioPrice * groupRatio,
+              cacheRatio: cacheRatio
+            })}</p>
+          )}
           <p></p>
           <p>
-            {i18next.t('提示 {{input}} tokens / 1M tokens * ${{price}} + 补全 {{completion}} tokens / 1M tokens * ${{compPrice}} * 分组 {{ratio}} = ${{total}}', {
-              input: inputTokens,
-              price: inputRatioPrice,
-              completion: completionTokens,
-              compPrice: completionRatioPrice,
-              ratio: groupRatio,
-              total: price.toFixed(6)
-            })}
+            {cacheTokens > 0 ? 
+              i18next.t('提示 {{nonCacheInput}} tokens + 缓存 {{cacheInput}} tokens * {{cacheRatio}} / 1M tokens * ${{price}} + 补全 {{completion}} tokens / 1M tokens * ${{compPrice}} * 分组 {{ratio}} = ${{total}}', {
+                nonCacheInput: inputTokens - cacheTokens,
+                cacheInput: cacheTokens,
+                cacheRatio: cacheRatio,
+                price: inputRatioPrice,
+                completion: completionTokens,
+                compPrice: completionRatioPrice,
+                ratio: groupRatio,
+                total: price.toFixed(6)
+              }) :
+              i18next.t('提示 {{input}} tokens / 1M tokens * ${{price}} + 补全 {{completion}} tokens / 1M tokens * ${{compPrice}} * 分组 {{ratio}} = ${{total}}', {
+                input: inputTokens,
+                price: inputRatioPrice,
+                completion: completionTokens,
+                compPrice: completionRatioPrice,
+                ratio: groupRatio,
+                total: price.toFixed(6)
+              })
+            }
           </p>
           <p>{i18next.t('仅供参考，以实际扣费为准')}</p>
         </article>
@@ -349,6 +377,8 @@ export function renderModelPriceSimple(
   modelRatio,
   modelPrice = -1,
   groupRatio,
+  cacheTokens = 0,
+  cacheRatio = 1.0,
 ) {
   if (modelPrice !== -1) {
     return i18next.t('价格：${{price}} * 分组：{{ratio}}', {
@@ -356,10 +386,18 @@ export function renderModelPriceSimple(
       ratio: groupRatio
     });
   } else {
-    return i18next.t('模型: {{ratio}} * 分组: {{groupRatio}}', {
-      ratio: modelRatio,
-      groupRatio: groupRatio
-    });
+    if (cacheTokens !== 0) {
+      return i18next.t('模型: {{ratio}} * 分组: {{groupRatio}} * 缓存比例: {{cacheRatio}}', {
+        ratio: modelRatio,
+        groupRatio: groupRatio,
+        cacheRatio: cacheRatio
+      });
+    } else {
+      return i18next.t('模型: {{ratio}} * 分组: {{groupRatio}}', {
+        ratio: modelRatio,
+        groupRatio: groupRatio
+      });
+    }
   }
 }
 
@@ -374,6 +412,8 @@ export function renderAudioModelPrice(
   audioRatio,
   audioCompletionRatio,
   groupRatio,
+  cacheTokens = 0,
+  cacheRatio = 1.0,
 ) {
   // 1 ratio = $0.002 / 1K tokens
   if (modelPrice !== -1) {
@@ -388,8 +428,13 @@ export function renderAudioModelPrice(
     // 这里的 *2 是因为 1倍率=0.002刀，请勿删除
     let inputRatioPrice = modelRatio * 2.0;
     let completionRatioPrice = modelRatio * 2.0 * completionRatio;
+    let cacheRatioPrice = modelRatio * 2.0 * cacheRatio;
+    
+    // Calculate effective input tokens (non-cached + cached with ratio applied)
+    const effectiveInputTokens = (inputTokens - cacheTokens) + (cacheTokens * cacheRatio);
+    
     let price =
-      (inputTokens / 1000000) * inputRatioPrice * groupRatio +
+      (effectiveInputTokens / 1000000) * inputRatioPrice * groupRatio +
       (completionTokens / 1000000) * completionRatioPrice * groupRatio +
       (audioInputTokens / 1000000) * inputRatioPrice * audioRatio * groupRatio +
       (audioCompletionTokens / 1000000) * inputRatioPrice * audioRatio * audioCompletionRatio * groupRatio;
@@ -406,6 +451,14 @@ export function renderAudioModelPrice(
             ratio: groupRatio,
             total: completionRatioPrice * groupRatio
           })}</p>
+          {cacheTokens > 0 && (
+            <p>{i18next.t('缓存：${{price}} * {{ratio}} = ${{total}} / 1M tokens (缓存比例: {{cacheRatio}})', {
+              price: cacheRatioPrice,
+              ratio: groupRatio,
+              total: cacheRatioPrice * groupRatio,
+              cacheRatio: cacheRatio
+            })}</p>
+          )}
           <p>{i18next.t('音频提示：${{price}} * {{ratio}} * {{audioRatio}} = ${{total}} / 1M tokens', {
             price: inputRatioPrice,
             ratio: groupRatio,
@@ -420,12 +473,22 @@ export function renderAudioModelPrice(
             total: inputRatioPrice * audioRatio * audioCompletionRatio * groupRatio
           })}</p>
           <p>
-            {i18next.t('文字提示 {{input}} tokens / 1M tokens * ${{price}} + 文字补全 {{completion}} tokens / 1M tokens * ${{compPrice}} +', {
-              input: inputTokens,
-              price: inputRatioPrice,
-              completion: completionTokens,
-              compPrice: completionRatioPrice
-            })}
+            {cacheTokens > 0 ? 
+              i18next.t('文字提示 {{nonCacheInput}} tokens + 文字缓存 {{cacheInput}} tokens * {{cacheRatio}} / 1M tokens * ${{price}} + 文字补全 {{completion}} tokens / 1M tokens * ${{compPrice}} +', {
+                nonCacheInput: inputTokens - cacheTokens,
+                cacheInput: cacheTokens,
+                cacheRatio: cacheRatio,
+                price: inputRatioPrice,
+                completion: completionTokens,
+                compPrice: completionRatioPrice
+              }) :
+              i18next.t('文字提示 {{input}} tokens / 1M tokens * ${{price}} + 文字补全 {{completion}} tokens / 1M tokens * ${{compPrice}} +', {
+                input: inputTokens,
+                price: inputRatioPrice,
+                completion: completionTokens,
+                compPrice: completionRatioPrice
+              })
+            }
           </p>
           <p>
             {i18next.t('音频提示 {{input}} tokens / 1M tokens * ${{price}} * {{audioRatio}} + 音频补全 {{completion}} tokens / 1M tokens * ${{price}} * {{audioRatio}} * {{audioCompRatio}}', {

web/src/i18n/locales/en.json

@@ -621,6 +621,7 @@
   "窗口等待": "window wait",
   "失败": "Failed",
   "绘图": "Drawing",
+  "绘图日志": "Drawing log",
   "放大": "Upscalers",
   "微妙放大": "Upscale (Subtle)",
   "创造放大": "Upscale (Creative)",
@@ -1074,10 +1075,9 @@
   "删除所选通道": "Delete selected channels",
   "标签聚合模式": "Enable tag mode",
   "没有账户？": "No account? ",
-  "注意，模型部署名称必须和模型名称保持一致，因为 One API 会把请求体中的 model 参数替换为你的部署名称（模型名称中的点会被剔除）": "Note: The model deployment name must match the model name because One API will replace the model parameter in the request body with your deployment name (dots in the model name will be removed)",
   "请输入 AZURE_OPENAI_ENDPOINT，例如：https://docs-test-001.openai.azure.com": "Please enter AZURE_OPENAI_ENDPOINT, e.g.: https://docs-test-001.openai.azure.com",
   "默认 API 版本": "Default API Version",
-  "请输入默认 API 版本，例如：2023-06-01-preview，该配置可以被实际的请求查询参数所覆盖": "Please enter default API version, e.g.: 2023-06-01-preview. This configuration can be overridden by actual request query parameters",
+  "请输入默认 API 版本，例如：2024-12-01-preview": "Please enter default API version, e.g.: 2024-12-01-preview.",
   "请为渠道命名": "Please name the channel",
   "请选择可以使用该渠道的分组": "Please select groups that can use this channel",
   "请在系统设置页面编辑分组倍率以添加新的分组：": "Please edit Group ratios in system settings to add new groups:",
@@ -1121,7 +1121,7 @@
   "知识库 ID": "Knowledge Base ID",
   "请输入知识库 ID，例如：123456": "Please enter knowledge base ID, e.g.: 123456",
   "可选值": "Optional value",
-  "异步任务": "Async task",
+  "任务日志": "Task log",
   "你好": "Hello",
   "你好，请问有什么可以帮助您的吗？": "Hello, how may I help you?",
   "用户分组": "Your default group",
@@ -1281,5 +1281,62 @@
   "频率限制的周期（分钟）": "Rate limit period (minutes)",
   "只包括请求成功的次数": "Only include successful request times",
   "保存模型速率限制": "Save model rate limit settings",
-  "速率限制设置": "Rate limit settings"
+  "速率限制设置": "Rate limit settings",
+  "获取启用模型失败:": "Failed to get enabled models:",
+  "获取启用模型失败": "Failed to get enabled models",
+  "JSON解析错误:": "JSON parsing error:",
+  "保存失败:": "Save failed:",
+  "输入模型倍率": "Enter model ratio",
+  "输入补全倍率": "Enter completion ratio",
+  "请输入数字": "Please enter a number",
+  "模型名称已存在": "Model name already exists",
+  "添加成功": "Added successfully",
+  "请先选择需要批量设置的模型": "Please select models for batch setting first",
+  "请输入模型倍率和补全倍率": "Please enter model ratio and completion ratio",
+  "请输入有效的数字": "Please enter a valid number",
+  "请输入填充值": "Please enter a value",
+  "批量设置成功": "Batch setting successful",
+  "已为 {{count}} 个模型设置{{type}}": "Set {{type}} for {{count}} models",
+  "固定价格": "Fixed Price",
+  "模型倍率和补全倍率": "Model Ratio and Completion Ratio",
+  "批量设置": "Batch Setting",
+  "搜索模型名称": "Search model name",
+  "此页面仅显示未设置价格或倍率的模型，设置后将自动从列表中移除": "This page only shows models without price or ratio settings. After setting, they will be automatically removed from the list",
+  "没有未设置的模型": "No unconfigured models",
+  "定价模式": "Pricing Mode",
+  "固定价格(每次)": "Fixed Price (per use)",
+  "输入每次价格": "Enter per-use price",
+  "批量设置模型参数": "Batch Set Model Parameters",
+  "设置类型": "Setting Type",
+  "模型倍率值": "Model Ratio Value",
+  "补全倍率值": "Completion Ratio Value",
+  "请输入模型倍率": "Enter model ratio",
+  "请输入补全倍率": "Enter completion ratio",
+  "请输入数值": "Enter a value",
+  "将为选中的 ": "Will set for selected ",
+  " 个模型设置相同的值": " models with the same value",
+  "当前设置类型: ": "Current setting type: ",
+  "固定价格值": "Fixed Price Value",
+  "未设置倍率模型": "Models without ratio settings",
+  "模型倍率和补全倍率同时设置": "Both model ratio and completion ratio are set",
+  "自用模式": "Self-use mode",
+  "开启后不限制：必须设置模型倍率": "After enabling, no limit: must set model ratio",
+  "演示站点模式": "Demo site mode",
+  "当前版本": "Current version",
+  "Gemini设置": "Gemini settings",
+  "Gemini安全设置": "Gemini safety settings",
+  "default为默认设置，可单独设置每个分类的安全等级": "\"default\" is the default setting, and each category can be set separately",
+  "Gemini版本设置": "Gemini version settings",
+  "default为默认设置，可单独设置每个模型的版本": "\"default\" is the default setting, and each model can be set separately",
+  "Claude设置": "Claude settings",
+  "Claude请求头覆盖": "Claude request header override",
+  "示例": "Example",
+  "缺省 MaxTokens": "Default MaxTokens",
+  "启用Claude思考适配（-thinking后缀）": "Enable Claude thinking adaptation (-thinking suffix)",
+  "Claude思考适配 BudgetTokens = MaxTokens * BudgetTokens 百分比": "Claude thinking adaptation BudgetTokens = MaxTokens * BudgetTokens percentage",
+  "思考适配 BudgetTokens 百分比": "Thinking adaptation BudgetTokens percentage",
+  "0.1-1之间的小数": "Decimal between 0.1 and 1",
+  "模型相关设置": "Model related settings",
+  "收起侧边栏": "Collapse sidebar",
+  "展开侧边栏": "Expand sidebar"
 }

web/src/pages/Channel/EditChannel.js

@@ -327,9 +327,6 @@ const EditChannel = (props) => {
         localInputs.base_url.length - 1
       );
     }
-    if (localInputs.type === 3 && localInputs.other === '') {
-      localInputs.other = '2023-06-01-preview';
-    }
     if (localInputs.type === 18 && localInputs.other === '') {
       localInputs.other = 'v2.1';
     }
@@ -494,7 +491,7 @@ const EditChannel = (props) => {
               <Input
                 label={t('默认 API 版本')}
                 name="azure_other"
-                placeholder={t('请输入默认 API 版本，例如：2023-06-01-preview，该配置可以被实际的请求查询参数所覆盖')}
+                placeholder={t('请输入默认 API 版本，例如：2024-12-01-preview')}
                 onChange={(value) => {
                   handleInputChange('other', value);
                 }}

web/src/pages/Setting/Model/SettingClaudeModel.js

@@ -0,0 +1,171 @@
+import React, { useEffect, useState, useRef } from 'react';
+import { Button, Col, Form, Row, Spin } from '@douyinfe/semi-ui';
+import {
+  compareObjects,
+  API,
+  showError,
+  showSuccess,
+  showWarning, verifyJSON
+} from '../../../helpers';
+import { useTranslation } from 'react-i18next';
+import Text from '@douyinfe/semi-ui/lib/es/typography/text';
+
+const CLAUDE_HEADER = {
+  'claude-3-7-sonnet-20250219-thinking': {
+    'anthropic-beta': ['output-128k-2025-02-19', 'token-efficient-tools-2025-02-19'],
+  }
+};
+
+const CLAUDE_DEFAULT_MAX_TOKENS = {
+  'default': 8192,
+  "claude-3-haiku-20240307": 4096,
+  "claude-3-opus-20240229": 4096,
+  'claude-3-7-sonnet-20250219-thinking': 8192,
+}
+
+export default function SettingClaudeModel(props) {
+  const { t } = useTranslation();
+
+  const [loading, setLoading] = useState(false);
+  const [inputs, setInputs] = useState({
+    'claude.model_headers_settings': '',
+    'claude.thinking_adapter_enabled': true,
+    'claude.default_max_tokens': '',
+    'claude.thinking_adapter_budget_tokens_percentage': 0.8,
+  });
+  const refForm = useRef();
+  const [inputsRow, setInputsRow] = useState(inputs);
+
+  function onSubmit() {
+    const updateArray = compareObjects(inputs, inputsRow);
+    if (!updateArray.length) return showWarning(t('你似乎并没有修改什么'));
+    const requestQueue = updateArray.map((item) => {
+      let value = String(inputs[item.key]);
+      
+      return API.put('/api/option/', {
+        key: item.key,
+        value,
+      });
+    });
+    setLoading(true);
+    Promise.all(requestQueue)
+      .then((res) => {
+        if (requestQueue.length === 1) {
+          if (res.includes(undefined)) return;
+        } else if (requestQueue.length > 1) {
+          if (res.includes(undefined)) return showError(t('部分保存失败，请重试'));
+        }
+        showSuccess(t('保存成功'));
+        props.refresh();
+      })
+      .catch(() => {
+        showError(t('保存失败，请重试'));
+      })
+      .finally(() => {
+        setLoading(false);
+      });
+  }
+
+  useEffect(() => {
+    const currentInputs = {};
+    for (let key in props.options) {
+      if (Object.keys(inputs).includes(key)) {
+        currentInputs[key] = props.options[key];
+      }
+    }
+    setInputs(currentInputs);
+    setInputsRow(structuredClone(currentInputs));
+    refForm.current.setValues(currentInputs);
+  }, [props.options]);
+
+  return (
+    <>
+      <Spin spinning={loading}>
+        <Form
+          values={inputs}
+          getFormApi={(formAPI) => (refForm.current = formAPI)}
+          style={{ marginBottom: 15 }}
+        >
+          <Form.Section text={t('Claude设置')}>
+            <Row>
+              <Col span={16}>
+                <Form.TextArea
+                  label={t('Claude请求头覆盖')}
+                  field={'claude.model_headers_settings'}
+                  placeholder={t('为一个 JSON 文本，例如：') + '\n' + JSON.stringify(CLAUDE_HEADER, null, 2)}
+                  extraText={t('示例') + '\n' + JSON.stringify(CLAUDE_HEADER, null, 2)}
+                  autosize={{ minRows: 6, maxRows: 12 }}
+                  trigger='blur'
+                  stopValidateWithError
+                  rules={[
+                    {
+                      validator: (rule, value) => verifyJSON(value),
+                      message: t('不是合法的 JSON 字符串')
+                    }
+                  ]}
+                  onChange={(value) => setInputs({ ...inputs, 'claude.model_headers_settings': value })}
+                />
+              </Col>
+            </Row>
+            <Row>
+              <Col span={8}>
+                <Form.TextArea
+                  label={t('缺省 MaxTokens')}
+                  field={'claude.default_max_tokens'}
+                  placeholder={t('为一个 JSON 文本，例如：') + '\n' + JSON.stringify(CLAUDE_DEFAULT_MAX_TOKENS, null, 2)}
+                  extraText={t('示例') + '\n' + JSON.stringify(CLAUDE_DEFAULT_MAX_TOKENS, null, 2)}
+                  autosize={{ minRows: 6, maxRows: 12 }}
+                  trigger='blur'
+                  stopValidateWithError
+                  rules={[
+                    {
+                      validator: (rule, value) => verifyJSON(value),
+                      message: t('不是合法的 JSON 字符串')
+                    }
+                  ]}
+                  onChange={(value) => setInputs({ ...inputs, 'claude.default_max_tokens': value })}
+                />
+              </Col>
+            </Row>
+            <Row>
+              <Col span={16}>
+                <Form.Switch
+                  label={t('启用Claude思考适配（-thinking后缀）')}
+                  field={'claude.thinking_adapter_enabled'}
+                  onChange={(value) => setInputs({ ...inputs, 'claude.thinking_adapter_enabled': value })}
+                />
+              </Col>
+            </Row>
+            <Row>
+              <Col span={16}>
+                {/*//展示MaxTokens和BudgetTokens的计算公式, 并展示实际数字*/}
+                <Text>
+                  {t('Claude思考适配 BudgetTokens = MaxTokens * BudgetTokens 百分比')}
+                </Text>
+              </Col>
+            </Row>
+            <Row>
+              <Col span={8}>
+                <Form.InputNumber
+                  label={t('思考适配 BudgetTokens 百分比')}
+                  field={'claude.thinking_adapter_budget_tokens_percentage'}
+                  initValue={''}
+                  extraText={t('0.1-1之间的小数')}
+                  min={0.1}
+                  max={1}
+                  onChange={(value) => setInputs({ ...inputs, 'claude.thinking_adapter_budget_tokens_percentage': value })}
+                />
+              </Col>
+            </Row>
+
+            <Row>
+              <Button size='default' onClick={onSubmit}>
+                {t('保存')}
+              </Button>
+            </Row>
+          </Form.Section>
+        </Form>
+      </Spin>
+    </>
+  );
+}

web/src/pages/Setting/Model/SettingGeminiModel.js

@@ -14,12 +14,18 @@ const GEMINI_SETTING_EXAMPLE = {
   'HARM_CATEGORY_CIVIC_INTEGRITY': 'BLOCK_NONE',
 };
 
+const GEMINI_VERSION_EXAMPLE = {
+  'default': 'v1beta',
+};
+
+
 export default function SettingGeminiModel(props) {
   const { t } = useTranslation();
 
   const [loading, setLoading] = useState(false);
   const [inputs, setInputs] = useState({
-    GeminiSafetySettings: '',
+    'gemini.safety_settings': '',
+    'gemini.version_settings': '',
   });
   const refForm = useRef();
   const [inputsRow, setInputsRow] = useState(inputs);
@@ -84,7 +90,7 @@ export default function SettingGeminiModel(props) {
                 <Form.TextArea
                   label={t('Gemini安全设置')}
                   placeholder={t('为一个 JSON 文本，例如：') + '\n' + JSON.stringify(GEMINI_SETTING_EXAMPLE, null, 2)}
-                  field={'GeminiSafetySettings'}
+                  field={'gemini.safety_settings'}
                   extraText={t('default为默认设置，可单独设置每个分类的安全等级')}
                   autosize={{ minRows: 6, maxRows: 12 }}
                   trigger='blur'
@@ -95,10 +101,31 @@ export default function SettingGeminiModel(props) {
                       message: t('不是合法的 JSON 字符串')
                     }
                   ]}
-                  onChange={(value) => setInputs({ ...inputs, GeminiSafetySettings: value })}
+                  onChange={(value) => setInputs({ ...inputs, 'gemini.safety_settings': value })}
                 />
               </Col>
             </Row>
+            <Row>
+              <Col span={16}>
+                <Form.TextArea
+                  label={t('Gemini版本设置')}
+                  placeholder={t('为一个 JSON 文本，例如：') + '\n' + JSON.stringify(GEMINI_VERSION_EXAMPLE, null, 2)}
+                  field={'gemini.version_settings'}
+                  extraText={t('default为默认设置，可单独设置每个模型的版本')}
+                  autosize={{ minRows: 6, maxRows: 12 }}
+                  trigger='blur'
+                  stopValidateWithError
+                  rules={[
+                    {
+                      validator: (rule, value) => verifyJSON(value),
+                      message: t('不是合法的 JSON 字符串')
+                    }
+                  ]}
+                  onChange={(value) => setInputs({ ...inputs, 'gemini.version_settings': value })}
+                />
+              </Col>
+            </Row>
+
             <Row>
               <Button size='default' onClick={onSubmit}>
                 {t('保存')}

web/src/pages/Setting/Operation/ModelRatioSettings.js

@@ -15,6 +15,7 @@ export default function ModelRatioSettings(props) {
   const [inputs, setInputs] = useState({
     ModelPrice: '',
     ModelRatio: '',
+    CacheRatio: '',
     CompletionRatio: '',
   });
   const refForm = useRef();
@@ -139,6 +140,25 @@ export default function ModelRatioSettings(props) {
               />
             </Col>
           </Row>
+          <Row gutter={16}>
+            <Col span={16}>
+              <Form.TextArea
+                label={t('提示缓存倍率')}
+                placeholder={t('为一个 JSON 文本，键为模型名称，值为倍率')}
+                field={'CacheRatio'}
+                autosize={{ minRows: 6, maxRows: 12 }}
+                trigger='blur'
+                stopValidateWithError
+                rules={[
+                  {
+                    validator: (rule, value) => verifyJSON(value),
+                    message: '不是合法的 JSON 字符串'
+                  }
+                ]}
+                onChange={(value) => setInputs({ ...inputs, CacheRatio: value })}
+              />
+            </Col>
+          </Row>
           <Row gutter={16}>
             <Col span={16}>
               <Form.TextArea

web/src/pages/Setting/Operation/ModelRationNotSetEditor.js

@@ -0,0 +1,549 @@
+import React, { useEffect, useState } from 'react';
+import { Table, Button, Input, Modal, Form, Space, Typography, Radio, Notification } from '@douyinfe/semi-ui';
+import { IconDelete, IconPlus, IconSearch, IconSave, IconBolt } from '@douyinfe/semi-icons';
+import { showError, showSuccess } from '../../../helpers';
+import { API } from '../../../helpers';
+import { useTranslation } from 'react-i18next';
+
+export default function ModelRatioNotSetEditor(props) {
+  const { t } = useTranslation();
+  const [models, setModels] = useState([]);
+  const [visible, setVisible] = useState(false);
+  const [batchVisible, setBatchVisible] = useState(false);
+  const [currentModel, setCurrentModel] = useState(null);
+  const [searchText, setSearchText] = useState('');
+  const [currentPage, setCurrentPage] = useState(1);
+  const [pageSize, setPageSize] = useState(10);
+  const [loading, setLoading] = useState(false);
+  const [enabledModels, setEnabledModels] = useState([]);
+  const [selectedRowKeys, setSelectedRowKeys] = useState([]);
+  const [batchFillType, setBatchFillType] = useState('ratio');
+  const [batchFillValue, setBatchFillValue] = useState('');
+  const [batchRatioValue, setBatchRatioValue] = useState('');
+  const [batchCompletionRatioValue, setBatchCompletionRatioValue] = useState('');
+  const { Text } = Typography;
+  // 定义可选的每页显示条数
+  const pageSizeOptions = [10, 20, 50, 100];
+
+  const getAllEnabledModels = async () => {
+    try {
+      const res = await API.get('/api/channel/models_enabled');
+      const { success, message, data } = res.data;
+      if (success) {
+        setEnabledModels(data);
+      } else {
+        showError(message);
+      }
+    } catch (error) {
+      console.error(t('获取启用模型失败:'), error);
+      showError(t('获取启用模型失败'));
+    }
+  }
+
+  useEffect(() => {
+    // 获取所有启用的模型
+    getAllEnabledModels();
+  }, []);
+
+  useEffect(() => {
+    try {
+      const modelPrice = JSON.parse(props.options.ModelPrice || '{}');
+      const modelRatio = JSON.parse(props.options.ModelRatio || '{}');
+      const completionRatio = JSON.parse(props.options.CompletionRatio || '{}');
+
+      // 找出所有未设置价格和倍率的模型
+      const unsetModels = enabledModels.filter(modelName => {
+        const hasPrice = modelPrice[modelName] !== undefined;
+        const hasRatio = modelRatio[modelName] !== undefined;
+        
+        // 如果模型没有价格或者没有倍率设置，则显示
+        return !hasPrice && !hasRatio;
+      });
+
+      // 创建模型数据
+      const modelData = unsetModels.map(name => ({
+        name,
+        price: modelPrice[name] || '',
+        ratio: modelRatio[name] || '',
+        completionRatio: completionRatio[name] || ''
+      }));
+
+      setModels(modelData);
+      // 清空选择
+      setSelectedRowKeys([]);
+    } catch (error) {
+      console.error(t('JSON解析错误:'), error);
+    }
+  }, [props.options, enabledModels]);
+
+  // 首先声明分页相关的工具函数
+  const getPagedData = (data, currentPage, pageSize) => {
+    const start = (currentPage - 1) * pageSize;
+    const end = start + pageSize;
+    return data.slice(start, end);
+  };
+
+  // 处理页面大小变化
+  const handlePageSizeChange = (size) => {
+    setPageSize(size);
+    // 重新计算当前页，避免数据丢失
+    const totalPages = Math.ceil(filteredModels.length / size);
+    if (currentPage > totalPages) {
+      setCurrentPage(totalPages || 1);
+    }
+  };
+
+  // 在 return 语句之前，先处理过滤和分页逻辑
+  const filteredModels = models.filter(model =>
+    searchText ? model.name.toLowerCase().includes(searchText.toLowerCase()) : true
+  );
+
+  // 然后基于过滤后的数据计算分页数据
+  const pagedData = getPagedData(filteredModels, currentPage, pageSize);
+
+  const SubmitData = async () => {
+    setLoading(true);
+    const output = {
+      ModelPrice: JSON.parse(props.options.ModelPrice || '{}'),
+      ModelRatio: JSON.parse(props.options.ModelRatio || '{}'),
+      CompletionRatio: JSON.parse(props.options.CompletionRatio || '{}')
+    };
+
+    try {
+      // 数据转换 - 只处理已修改的模型
+      models.forEach(model => {
+        // 只有当用户设置了值时才更新
+        if (model.price !== '') {
+          // 如果价格不为空，则转换为浮点数，忽略倍率参数
+          output.ModelPrice[model.name] = parseFloat(model.price);
+        } else {
+          if (model.ratio !== '') output.ModelRatio[model.name] = parseFloat(model.ratio);
+          if (model.completionRatio !== '') output.CompletionRatio[model.name] = parseFloat(model.completionRatio);
+        }
+      });
+
+      // 准备API请求数组
+      const finalOutput = {
+        ModelPrice: JSON.stringify(output.ModelPrice, null, 2),
+        ModelRatio: JSON.stringify(output.ModelRatio, null, 2),
+        CompletionRatio: JSON.stringify(output.CompletionRatio, null, 2)
+      };
+
+      const requestQueue = Object.entries(finalOutput).map(([key, value]) => {
+        return API.put('/api/option/', {
+          key,
+          value
+        });
+      });
+
+      // 批量处理请求
+      const results = await Promise.all(requestQueue);
+
+      // 验证结果
+      if (requestQueue.length === 1) {
+        if (results.includes(undefined)) return;
+      } else if (requestQueue.length > 1) {
+        if (results.includes(undefined)) {
+          return showError(t('部分保存失败，请重试'));
+        }
+      }
+
+      // 检查每个请求的结果
+      for (const res of results) {
+        if (!res.data.success) {
+          return showError(res.data.message);
+        }
+      }
+
+      showSuccess(t('保存成功'));
+      props.refresh();
+      // 重新获取未设置的模型
+      getAllEnabledModels();
+
+    } catch (error) {
+      console.error(t('保存失败:'), error);
+      showError(t('保存失败，请重试'));
+    } finally {
+      setLoading(false);
+    }
+  };
+
+  const columns = [
+    {
+      title: t('模型名称'),
+      dataIndex: 'name',
+      key: 'name',
+    },
+    {
+      title: t('模型固定价格'),
+      dataIndex: 'price',
+      key: 'price',
+      render: (text, record) => (
+        <Input
+          value={text}
+          placeholder={t('按量计费')}
+          onChange={value => updateModel(record.name, 'price', value)}
+        />
+      )
+    },
+    {
+      title: t('模型倍率'),
+      dataIndex: 'ratio',
+      key: 'ratio',
+      render: (text, record) => (
+        <Input
+          value={text}
+          placeholder={record.price !== '' ? t('模型倍率') : t('输入模型倍率')}
+          disabled={record.price !== ''}
+          onChange={value => updateModel(record.name, 'ratio', value)}
+        />
+      )
+    },
+    {
+      title: t('补全倍率'),
+      dataIndex: 'completionRatio',
+      key: 'completionRatio',
+      render: (text, record) => (
+        <Input
+          value={text}
+          placeholder={record.price !== '' ? t('补全倍率') : t('输入补全倍率')}
+          disabled={record.price !== ''}
+          onChange={value => updateModel(record.name, 'completionRatio', value)}
+        />
+      )
+    }
+  ];
+
+  const updateModel = (name, field, value) => {
+    if (value !== '' && isNaN(value)) {
+      showError(t('请输入数字'));
+      return;
+    }
+    setModels(prev =>
+      prev.map(model =>
+        model.name === name
+          ? { ...model, [field]: value }
+          : model
+      )
+    );
+  };
+
+  const addModel = (values) => {
+    // 检查模型名称是否存在, 如果存在则拒绝添加
+    if (models.some(model => model.name === values.name)) {
+      showError(t('模型名称已存在'));
+      return;
+    }
+    setModels(prev => [{
+      name: values.name,
+      price: values.price || '',
+      ratio: values.ratio || '',
+      completionRatio: values.completionRatio || ''
+    }, ...prev]);
+    setVisible(false);
+    showSuccess(t('添加成功'));
+  };
+
+  // 批量填充功能
+  const handleBatchFill = () => {
+    if (selectedRowKeys.length === 0) {
+      showError(t('请先选择需要批量设置的模型'));
+      return;
+    }
+
+    if (batchFillType === 'bothRatio') {
+      if (batchRatioValue === '' || batchCompletionRatioValue === '') {
+        showError(t('请输入模型倍率和补全倍率'));
+        return;
+      }
+      if (isNaN(batchRatioValue) || isNaN(batchCompletionRatioValue)) {
+        showError(t('请输入有效的数字'));
+        return;
+      }
+    } else {
+      if (batchFillValue === '') {
+        showError(t('请输入填充值'));
+        return;
+      }
+      if (isNaN(batchFillValue)) {
+        showError(t('请输入有效的数字'));
+        return;
+      }
+    }
+
+    // 根据选择的类型批量更新模型
+    setModels(prev => 
+      prev.map(model => {
+        if (selectedRowKeys.includes(model.name)) {
+          if (batchFillType === 'price') {
+            return {
+              ...model,
+              price: batchFillValue,
+              ratio: '',
+              completionRatio: ''
+            };
+          } else if (batchFillType === 'ratio') {
+            return {
+              ...model,
+              price: '',
+              ratio: batchFillValue
+            };
+          } else if (batchFillType === 'completionRatio') {
+            return {
+              ...model,
+              price: '',
+              completionRatio: batchFillValue
+            };
+          } else if (batchFillType === 'bothRatio') {
+            return {
+              ...model,
+              price: '',
+              ratio: batchRatioValue,
+              completionRatio: batchCompletionRatioValue
+            };
+          }
+        }
+        return model;
+      })
+    );
+
+    setBatchVisible(false);
+    Notification.success({
+      title: t('批量设置成功'),
+      content: t('已为 {{count}} 个模型设置{{type}}', {
+        count: selectedRowKeys.length,
+        type: batchFillType === 'price' ? t('固定价格') : 
+              batchFillType === 'ratio' ? t('模型倍率') : 
+              batchFillType === 'completionRatio' ? t('补全倍率') : t('模型倍率和补全倍率')
+      }),
+      duration: 3,
+    });
+  };
+
+  const handleBatchTypeChange = (value) => {
+    console.log(t('Changing batch type to:'), value);
+    setBatchFillType(value);
+    
+    // 切换类型时清空对应的值
+    if (value !== 'bothRatio') {
+      setBatchFillValue('');
+    } else {
+      setBatchRatioValue('');
+      setBatchCompletionRatioValue('');
+    }
+  };
+
+  const rowSelection = {
+    selectedRowKeys,
+    onChange: (selectedKeys) => {
+      setSelectedRowKeys(selectedKeys);
+    },
+  };
+
+  return (
+    <>
+      <Space vertical align="start" style={{ width: '100%' }}>
+        <Space>
+          <Button icon={<IconPlus />} onClick={() => setVisible(true)}>
+            {t('添加模型')}
+          </Button>
+          <Button 
+            icon={<IconBolt />} 
+            type="secondary"
+            onClick={() => setBatchVisible(true)}
+            disabled={selectedRowKeys.length === 0}
+          >
+            {t('批量设置')} ({selectedRowKeys.length})
+          </Button>
+          <Button type="primary" icon={<IconSave />} onClick={SubmitData} loading={loading}>
+            {t('应用更改')}
+          </Button>
+          <Input
+            prefix={<IconSearch />}
+            placeholder={t('搜索模型名称')}
+            value={searchText}
+            onChange={value => {
+              setSearchText(value)
+              setCurrentPage(1);
+            }}
+            style={{ width: 200 }}
+          />
+        </Space>
+
+        <Text>{t('此页面仅显示未设置价格或倍率的模型，设置后将自动从列表中移除')}</Text>
+        
+        <Table
+          columns={columns}
+          dataSource={pagedData}
+          rowSelection={rowSelection}
+          rowKey="name"
+          pagination={{
+            currentPage: currentPage,
+            pageSize: pageSize,
+            total: filteredModels.length,
+            onPageChange: page => setCurrentPage(page),
+            onPageSizeChange: handlePageSizeChange,
+            pageSizeOptions: pageSizeOptions,
+            formatPageText: (page) =>
+              t('第 {{start}} - {{end}} 条，共 {{total}} 条', {
+                start: page.currentStart,
+                end: page.currentEnd,
+                total: filteredModels.length
+              }),
+            showTotal: true,
+            showSizeChanger: true
+          }}
+          empty={
+            <div style={{ textAlign: 'center', padding: '20px' }}>
+              {t('没有未设置的模型')}
+            </div>
+          }
+        />
+      </Space>
+
+      {/* 添加模型弹窗 */}
+      <Modal
+        title={t('添加模型')}
+        visible={visible}
+        onCancel={() => setVisible(false)}
+        onOk={() => {
+          currentModel && addModel(currentModel);
+        }}
+      >
+        <Form>
+          <Form.Input
+            field="name"
+            label={t('模型名称')}
+            placeholder="strawberry"
+            required
+            onChange={value => setCurrentModel(prev => ({ ...prev, name: value }))}
+          />
+          <Form.Switch
+            field="priceMode"
+            label={<>{t('定价模式')}：{currentModel?.priceMode ? t("固定价格") : t("倍率模式")}</>}
+            onChange={checked => {
+              setCurrentModel(prev => ({
+                ...prev,
+                price: '',
+                ratio: '',
+                completionRatio: '',
+                priceMode: checked
+              }));
+            }}
+          />
+          {currentModel?.priceMode ? (
+            <Form.Input
+              field="price"
+              label={t('固定价格(每次)')}
+              placeholder={t('输入每次价格')}
+              onChange={value => setCurrentModel(prev => ({ ...prev, price: value }))}
+            />
+          ) : (
+            <>
+              <Form.Input
+                field="ratio"
+                label={t('模型倍率')}
+                placeholder={t('输入模型倍率')}
+                onChange={value => setCurrentModel(prev => ({ ...prev, ratio: value }))}
+              />
+              <Form.Input
+                field="completionRatio"
+                label={t('补全倍率')}
+                placeholder={t('输入补全价格')}
+                onChange={value => setCurrentModel(prev => ({ ...prev, completionRatio: value }))}
+              />
+            </>
+          )}
+        </Form>
+      </Modal>
+
+      {/* 批量设置弹窗 */}
+      <Modal
+        title={t('批量设置模型参数')}
+        visible={batchVisible}
+        onCancel={() => setBatchVisible(false)}
+        onOk={handleBatchFill}
+        width={500}
+      >
+        <Form>
+          <Form.Section text={t('设置类型')}>
+            <div style={{ marginBottom: '16px' }}>
+              <Space>
+                <Radio
+                  checked={batchFillType === 'price'}
+                  onChange={() => handleBatchTypeChange('price')}
+                >
+                  {t('固定价格')}
+                </Radio>
+                <Radio
+                  checked={batchFillType === 'ratio'}
+                  onChange={() => handleBatchTypeChange('ratio')}
+                >
+                  {t('模型倍率')}
+                </Radio>
+                <Radio
+                  checked={batchFillType === 'completionRatio'}
+                  onChange={() => handleBatchTypeChange('completionRatio')}
+                >
+                  {t('补全倍率')}
+                </Radio>
+                <Radio
+                  checked={batchFillType === 'bothRatio'}
+                  onChange={() => handleBatchTypeChange('bothRatio')}
+                >
+                  {t('模型倍率和补全倍率同时设置')}
+                </Radio>
+              </Space>
+            </div>
+          </Form.Section>
+          
+          {batchFillType === 'bothRatio' ? (
+            <>
+              <Form.Input
+                field="batchRatioValue"
+                label={t('模型倍率值')}
+                placeholder={t('请输入模型倍率')}
+                value={batchRatioValue}
+                onChange={value => setBatchRatioValue(value)}
+              />
+              <Form.Input
+                field="batchCompletionRatioValue"
+                label={t('补全倍率值')}
+                placeholder={t('请输入补全倍率')}
+                value={batchCompletionRatioValue}
+                onChange={value => setBatchCompletionRatioValue(value)}
+              />
+            </>
+          ) : (
+            <Form.Input
+              field="batchFillValue"
+              label={
+                batchFillType === 'price' 
+                  ? t('固定价格值') 
+                  : batchFillType === 'ratio'
+                    ? t('模型倍率值')
+                    : t('补全倍率值')
+              }
+              placeholder={t('请输入数值')}
+              value={batchFillValue}
+              onChange={value => setBatchFillValue(value)}
+            />
+          )}
+          
+          <Text type="tertiary">
+            {t('将为选中的 ')} <Text strong>{selectedRowKeys.length}</Text> {t(' 个模型设置相同的值')}
+          </Text>
+          <div style={{ marginTop: '8px' }}>
+            <Text type="tertiary">
+              {t('当前设置类型: ')} <Text strong>{
+                batchFillType === 'price' ? t('固定价格') : 
+                batchFillType === 'ratio' ? t('模型倍率') : 
+                batchFillType === 'completionRatio' ? t('补全倍率') : t('模型倍率和补全倍率')
+              }</Text>
+            </Text>
+          </div>
+        </Form>
+      </Modal>
+    </>
+  );
+}

web/src/pages/Setting/Operation/SettingsGeneral.js

@@ -22,6 +22,7 @@ export default function GeneralSettings(props) {
     DisplayTokenStatEnabled: false,
     DefaultCollapseSidebar: false,
     DemoSiteEnabled: false,
+    SelfUseModeEnabled: false,
   });
   const refForm = useRef();
   const [inputsRow, setInputsRow] = useState(inputs);
@@ -205,6 +206,22 @@ export default function GeneralSettings(props) {
                   }
                 />
               </Col>
+              <Col span={8}>
+                <Form.Switch
+                  field={'SelfUseModeEnabled'}
+                  label={t('自用模式')}
+                  extraText={t('开启后不限制：必须设置模型倍率')}
+                  size='default'
+                  checkedText='｜'
+                  uncheckedText='〇'
+                  onChange={(value) =>
+                    setInputs({
+                      ...inputs,
+                      SelfUseModeEnabled: value
+                    })
+                  }
+                />
+              </Col>
             </Row>
             <Row>
               <Button size='default' onClick={onSubmit}>

web/src/pages/Setting/index.js

@@ -18,9 +18,6 @@ const Setting = () => {
   const [tabActiveKey, setTabActiveKey] = useState('1');
   let panes = [
     {
-      tab: t('个人设置'),
-      content: <PersonalSetting />,
-      itemKey: 'personal',
     },
   ];
 
@@ -61,7 +58,7 @@ const Setting = () => {
     if (tab) {
       setTabActiveKey(tab);
     } else {
-      onChangeTab('personal');
+      onChangeTab('operation');
     }
   }, [location.search]);
   return (