Merge pull request #893 from wizcas/replace-linux-do-icon

替换登录界面的 Linux.do OAuth 图标

README.md

@@ -36,8 +36,8 @@
 > 本项目为开源项目，在[One API](https://github.com/songquanpeng/one-api)的基础上进行二次开发
 
 > [!IMPORTANT]  
-> - 使用者必须在遵循 OpenAI 的[使用条款](https://openai.com/policies/terms-of-use)以及**法律法规**的情况下使用，不得用于非法用途。
 > - 本项目仅供个人学习使用，不保证稳定性，且不提供任何技术支持。
+> - 使用者必须在遵循 OpenAI 的[使用条款](https://openai.com/policies/terms-of-use)以及**法律法规**的情况下使用，不得用于非法用途。
 > - 根据[《生成式人工智能服务管理暂行办法》](http://www.cac.gov.cn/2023-07/13/c_1690898327029107.htm)的要求，请勿对中国地区公众提供一切未经备案的生成式人工智能服务。
 
 ## 📚 文档
@@ -46,35 +46,32 @@
 
 ## ✨ 主要特性
 
-New API提供了丰富的功能，详细特性请参考[维基百科-特性说明](https://docs.newapi.pro/wiki/features-introduction)：
+New API提供了丰富的功能，详细特性请参考[特性说明](https://docs.newapi.pro/wiki/features-introduction)：
 
 1. 🎨 全新的UI界面
 2. 🌍 多语言支持
-3. 🎨 支持[Midjourney-Proxy(Plus)](https://github.com/novicezk/midjourney-proxy)接口，[对接文档](https://docs.newapi.pro/api/relay/image/midjourney)
-4. 💰 支持在线充值功能（易支付）
-5. 🔍 支持用key查询使用额度（配合[neko-api-key-tool](https://github.com/Calcium-Ion/neko-api-key-tool)）
-6. 📑 分页支持选择每页显示数量
-7. 🔄 兼容原版One API的数据库
-8. 💵 支持模型按次数收费
-9. ⚖️ 支持渠道加权随机
-10. 📈 数据看板（控制台）
-11. 🔒 可设置令牌能调用的模型
-12. 🤖 支持Telegram授权登录
-13. 🎵 支持[Suno API](https://github.com/Suno-API/Suno-API)接口，[接口文档](https://docs.newapi.pro/api/suno-music)
-14. 🔄 支持Rerank模型（Cohere和Jina），[接口文档](https://docs.newapi.pro/api/jinaai-rerank)
-15. ⚡ 支持OpenAI Realtime API（包括Azure渠道），[接口文档](https://docs.newapi.pro/api/openai-realtime)
-16. ⚡ 支持Claude Messages 格式，[接口文档](https://docs.newapi.pro/api/anthropic-chat)
-17. 支持使用路由/chat2link进入聊天界面
-18. 🧠 支持通过模型名称后缀设置 reasoning effort：
+3. 💰 支持在线充值功能（易支付）
+4. 🔍 支持用key查询使用额度（配合[neko-api-key-tool](https://github.com/Calcium-Ion/neko-api-key-tool)）
+5. 🔄 兼容原版One API的数据库
+6. 💵 支持模型按次数收费
+7. ⚖️ 支持渠道加权随机
+8. 📈 数据看板（控制台）
+9. 🔒 令牌分组、模型限制
+10. 🤖 支持更多授权登陆方式（LinuxDO,Telegram、OIDC）
+11. 🔄 支持Rerank模型（Cohere和Jina），[接口文档](https://docs.newapi.pro/api/jinaai-rerank)
+12. ⚡ 支持OpenAI Realtime API（包括Azure渠道），[接口文档](https://docs.newapi.pro/api/openai-realtime)
+13. ⚡ 支持Claude Messages 格式，[接口文档](https://docs.newapi.pro/api/anthropic-chat)
+14. 支持使用路由/chat2link进入聊天界面
+15. 🧠 支持通过模型名称后缀设置 reasoning effort：
     1. OpenAI o系列模型
         - 添加后缀 `-high` 设置为 high reasoning effort (例如: `o3-mini-high`)
         - 添加后缀 `-medium` 设置为 medium reasoning effort (例如: `o3-mini-medium`)
         - 添加后缀 `-low` 设置为 low reasoning effort (例如: `o3-mini-low`)
     2. Claude 思考模型
         - 添加后缀 `-thinking` 启用思考模式 (例如: `claude-3-7-sonnet-20250219-thinking`)
-19. 🔄 思考转内容功能
-20. 🔄 模型限流功能
-20. 💰 缓存计费支持，开启后可以在缓存命中时按照设定的比例计费：
+16. 🔄 思考转内容功能
+17. 🔄 针对用户的模型限流功能
+18. 💰 缓存计费支持，开启后可以在缓存命中时按照设定的比例计费：
     1. 在 `系统设置-运营设置` 中设置 `提示缓存倍率` 选项
     2. 在渠道中设置 `提示缓存倍率`，范围 0-1，例如设置为 0.5 表示缓存命中时按照 50% 计费
     3. 支持的渠道：
@@ -88,12 +85,12 @@ New API提供了丰富的功能，详细特性请参考[维基百科-特性说
 此版本支持多种模型，详情请参考[接口文档-中继接口](https://docs.newapi.pro/api)：
 
 1. 第三方模型 **gpts** （gpt-4-gizmo-*）
-2. [Midjourney-Proxy(Plus)](https://github.com/novicezk/midjourney-proxy)接口，[接口文档](https://docs.newapi.pro/api/midjourney-proxy-image)
-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，[接口文档](https://docs.newapi.pro/api/suno-music)
+2. 第三方渠道[Midjourney-Proxy(Plus)](https://github.com/novicezk/midjourney-proxy)接口，[接口文档](https://docs.newapi.pro/api/midjourney-proxy-image)
+3. 第三方渠道[Suno API](https://github.com/Suno-API/Suno-API)接口，[接口文档](https://docs.newapi.pro/api/suno-music)
+4. 自定义渠道，支持填入完整调用地址
 5. Rerank模型（[Cohere](https://cohere.ai/)和[Jina](https://jina.ai/)），[接口文档](https://docs.newapi.pro/api/jinaai-rerank)
 6. Claude Messages 格式，[接口文档](https://docs.newapi.pro/api/anthropic-chat)
-7. Dify
+7. Dify，当前仅支持chatflow
 
 ## 环境变量配置
 
@@ -168,9 +165,6 @@ docker run --name new-api -d --restart always -p 3000:3000 -e SQL_DSN="root:1234
 
 - [聊天接口（Chat）](https://docs.newapi.pro/api/openai-chat)
 - [图像接口（Image）](https://docs.newapi.pro/api/openai-image)
-- [Midjourney接口](https://docs.newapi.pro/api/midjourney-proxy-image)
-- [音乐接口（Music）](https://docs.newapi.pro/api/relay/music)
-- [Suno接口](https://docs.newapi.pro/api/suno-music)
 - [重排序接口（Rerank）](https://docs.newapi.pro/api/jinaai-rerank)
 - [实时对话接口（Realtime）](https://docs.newapi.pro/api/openai-realtime)
 - [Claude聊天接口（messages）](https://docs.newapi.pro/api/anthropic-chat)

controller/channel-test.go

@@ -187,7 +187,9 @@ func buildTestRequest(model string) *dto.GeneralOpenAIRequest {
 	if strings.HasPrefix(model, "o1") || strings.HasPrefix(model, "o3") {
 		testRequest.MaxCompletionTokens = 10
 	} else if strings.Contains(model, "thinking") {
-		testRequest.MaxTokens = 50
+		if !strings.Contains(model, "claude") {
+			testRequest.MaxTokens = 50
+		}
 	} else {
 		testRequest.MaxTokens = 10
 	}

controller/option.go

@@ -53,11 +53,12 @@ func UpdateOption(c *gin.Context) {
 			return
 		}
 	case "oidc.enabled":
-		if option.Value == "true" && system_setting.GetOIDCSettings().Enabled {
+		if option.Value == "true" && system_setting.GetOIDCSettings().ClientId == "" {
 			c.JSON(http.StatusOK, gin.H{
 				"success": false,
 				"message": "无法启用 OIDC 登录，请先填入 OIDC Client Id 以及 OIDC Client Secret！",
 			})
+			return
 		}
 	case "LinuxDOOAuthEnabled":
 		if option.Value == "true" && common.LinuxDOClientId == "" {
@@ -89,6 +90,15 @@ func UpdateOption(c *gin.Context) {
 				"success": false,
 				"message": "无法启用 Turnstile 校验，请先填入 Turnstile 校验相关配置信息！",
 			})
+
+			return
+		}
+	case "TelegramOAuthEnabled":
+		if option.Value == "true" && common.TelegramBotToken == "" {
+			c.JSON(http.StatusOK, gin.H{
+				"success": false,
+				"message": "无法启用 Telegram OAuth，请先填入 Telegram Bot Token！",
+			})
 			return
 		}
 	case "GroupRatio":
@@ -100,6 +110,7 @@ func UpdateOption(c *gin.Context) {
 			})
 			return
 		}
+
 	}
 	err = model.UpdateOption(option.Key, option.Value)
 	if err != nil {

docs/channel/other_setting.md

@@ -11,7 +11,7 @@
     - 类型为字符串，填写代理地址（例如 socks5 协议的代理地址）
 
 3. thinking_to_content
-   - 用于标识是否将思考内容`reasoning_conetnt`转换为`<think>`标签拼接到内容中返回
+   - 用于标识是否将思考内容`reasoning_content`转换为`<think>`标签拼接到内容中返回
    - 类型为布尔值，设置为 true 时启用思考内容转换
 
 --------------------------------------------------------------
@@ -30,4 +30,4 @@
 
 --------------------------------------------------------------
 
-通过调整上述 JSON 配置中的值，可以灵活控制渠道的额外行为，比如是否进行格式化以及使用特定的网络代理。
+通过调整上述 JSON 配置中的值，可以灵活控制渠道的额外行为，比如是否进行格式化以及使用特定的网络代理。

dto/claude.go

@@ -183,7 +183,7 @@ type ClaudeResponse struct {
 	Completion   string               `json:"completion,omitempty"`
 	StopReason   string               `json:"stop_reason,omitempty"`
 	Model        string               `json:"model,omitempty"`
-	Error        ClaudeError          `json:"error,omitempty"`
+	Error        *ClaudeError         `json:"error,omitempty"`
 	Usage        *ClaudeUsage         `json:"usage,omitempty"`
 	Index        *int                 `json:"index,omitempty"`
 	ContentBlock *ClaudeMediaMessage  `json:"content_block,omitempty"`

dto/rerank.go

@@ -5,18 +5,29 @@ type RerankRequest struct {
 	Query           string `json:"query"`
 	Model           string `json:"model"`
 	TopN            int    `json:"top_n"`
-	ReturnDocuments bool   `json:"return_documents,omitempty"`
+	ReturnDocuments *bool  `json:"return_documents,omitempty"`
 	MaxChunkPerDoc  int    `json:"max_chunk_per_doc,omitempty"`
 	OverLapTokens   int    `json:"overlap_tokens,omitempty"`
 }
 
-type RerankResponseDocument struct {
+func (r *RerankRequest) GetReturnDocuments() bool {
+	if r.ReturnDocuments == nil {
+		return false
+	}
+	return *r.ReturnDocuments
+}
+
+type RerankResponseResult struct {
 	Document       any     `json:"document,omitempty"`
 	Index          int     `json:"index"`
 	RelevanceScore float64 `json:"relevance_score"`
 }
 
-type RerankResponse struct {
-	Results []RerankResponseDocument `json:"results"`
-	Usage   Usage                    `json:"usage"`
+type RerankDocument struct {
+	Text any `json:"text"`
+}
+
+type RerankResponse struct {
+	Results []RerankResponseResult `json:"results"`
+	Usage   Usage                  `json:"usage"`
 }

middleware/distributor.go

@@ -212,6 +212,7 @@ func SetupContextForSelectedChannel(c *gin.Context, channel *model.Channel, mode
 	c.Set("channel_name", channel.Name)
 	c.Set("channel_type", channel.Type)
 	c.Set("channel_setting", channel.GetSetting())
+	c.Set("param_override", channel.GetParamOverride())
 	if nil != channel.OpenAIOrganization && "" != *channel.OpenAIOrganization {
 		c.Set("channel_organization", *channel.OpenAIOrganization)
 	}

model/channel.go

@@ -36,6 +36,7 @@ type Channel struct {
 	OtherInfo         string  `json:"other_info"`
 	Tag               *string `json:"tag" gorm:"index"`
 	Setting           *string `json:"setting" gorm:"type:text"`
+	ParamOverride     *string `json:"param_override" gorm:"type:text"`
 }
 
 func (channel *Channel) GetModels() []string {
@@ -511,6 +512,17 @@ func (channel *Channel) SetSetting(setting map[string]interface{}) {
 	channel.Setting = common.GetPointer[string](string(settingBytes))
 }
 
+func (channel *Channel) GetParamOverride() map[string]interface{} {
+	paramOverride := make(map[string]interface{})
+	if channel.ParamOverride != nil && *channel.ParamOverride != "" {
+		err := json.Unmarshal([]byte(*channel.ParamOverride), &paramOverride)
+		if err != nil {
+			common.SysError("failed to unmarshal param override: " + err.Error())
+		}
+	}
+	return paramOverride
+}
+
 func GetChannelsByIds(ids []int) ([]*Channel, error) {
 	var channels []*Channel
 	err := DB.Where("id in (?)", ids).Find(&channels).Error

relay/channel/aws/adaptor.go

@@ -21,6 +21,8 @@ type Adaptor struct {
 }
 
 func (a *Adaptor) ConvertClaudeRequest(c *gin.Context, info *relaycommon.RelayInfo, request *dto.ClaudeRequest) (any, error) {
+	c.Set("request_model", request.Model)
+	c.Set("converted_request", request)
 	return request, nil
 }
 

relay/channel/aws/constants.go

@@ -13,4 +13,41 @@ var awsModelIDMap = map[string]string{
 	"claude-3-7-sonnet-20250219": "anthropic.claude-3-7-sonnet-20250219-v1:0",
 }
 
+var awsModelCanCrossRegionMap = map[string]map[string]bool{
+	"anthropic.claude-3-sonnet-20240229-v1:0": {
+		"us": true,
+		"eu": true,
+		"ap": true,
+	},
+	"anthropic.claude-3-opus-20240229-v1:0": {
+		"us": true,
+	},
+	"anthropic.claude-3-haiku-20240307-v1:0": {
+		"us": true,
+		"eu": true,
+		"ap": true,
+	},
+	"anthropic.claude-3-5-sonnet-20240620-v1:0": {
+		"us": true,
+		"eu": true,
+		"ap": true,
+	},
+	"anthropic.claude-3-5-sonnet-20241022-v2:0": {
+		"us": true,
+		"ap": true,
+	},
+	"anthropic.claude-3-5-haiku-20241022-v1:0": {
+		"us": true,
+	},
+	"anthropic.claude-3-7-sonnet-20250219-v1:0": {
+		"us": true,
+	},
+}
+
+var awsRegionCrossModelPrefixMap = map[string]string{
+	"us": "us",
+	"eu": "eu",
+	"ap": "apac",
+}
+
 var ChannelName = "aws"

relay/channel/aws/relay-aws.go

@@ -43,6 +43,28 @@ func wrapErr(err error) *dto.OpenAIErrorWithStatusCode {
 	}
 }
 
+func awsRegionPrefix(awsRegionId string) string {
+	parts := strings.Split(awsRegionId, "-")
+	regionPrefix := ""
+	if len(parts) > 0 {
+		regionPrefix = parts[0]
+	}
+	return regionPrefix
+}
+
+func awsModelCanCrossRegion(awsModelId, awsRegionPrefix string) bool {
+	regionSet, exists := awsModelCanCrossRegionMap[awsModelId]
+	return exists && regionSet[awsRegionPrefix]
+}
+
+func awsModelCrossRegion(awsModelId, awsRegionPrefix string) string {
+	modelPrefix, find := awsRegionCrossModelPrefixMap[awsRegionPrefix]
+	if !find {
+		return awsModelId
+	}
+	return modelPrefix + "." + awsModelId
+}
+
 func awsModelID(requestModel string) (string, error) {
 	if awsModelID, ok := awsModelIDMap[requestModel]; ok {
 		return awsModelID, nil
@@ -62,6 +84,12 @@ func awsHandler(c *gin.Context, info *relaycommon.RelayInfo, requestMode int) (*
 		return wrapErr(errors.Wrap(err, "awsModelID")), nil
 	}
 
+	awsRegionPrefix := awsRegionPrefix(awsCli.Options().Region)
+	canCrossRegion := awsModelCanCrossRegion(awsModelId, awsRegionPrefix)
+	if canCrossRegion {
+		awsModelId = awsModelCrossRegion(awsModelId, awsRegionPrefix)
+	}
+
 	awsReq := &bedrockruntime.InvokeModelInput{
 		ModelId:     aws.String(awsModelId),
 		Accept:      aws.String("application/json"),
@@ -107,6 +135,12 @@ func awsStreamHandler(c *gin.Context, resp *http.Response, info *relaycommon.Rel
 		return wrapErr(errors.Wrap(err, "awsModelID")), nil
 	}
 
+	awsRegionPrefix := awsRegionPrefix(awsCli.Options().Region)
+	canCrossRegion := awsModelCanCrossRegion(awsModelId, awsRegionPrefix)
+	if canCrossRegion {
+		awsModelId = awsModelCrossRegion(awsModelId, awsRegionPrefix)
+	}
+
 	awsReq := &bedrockruntime.InvokeModelWithResponseStreamInput{
 		ModelId:     aws.String(awsModelId),
 		Accept:      aws.String("application/json"),

relay/channel/claude/relay-claude.go

@@ -70,7 +70,9 @@ func RequestOpenAI2ClaudeMessage(textRequest dto.GeneralOpenAIRequest) (*dto.Cla
 				Description: tool.Function.Description,
 			}
 			claudeTool.InputSchema = make(map[string]interface{})
-			claudeTool.InputSchema["type"] = params["type"].(string)
+			if params["type"] != nil {
+				claudeTool.InputSchema["type"] = params["type"].(string)
+			}
 			claudeTool.InputSchema["properties"] = params["properties"]
 			claudeTool.InputSchema["required"] = params["required"]
 			for s, a := range params {
@@ -485,7 +487,7 @@ func HandleStreamResponseData(c *gin.Context, info *relaycommon.RelayInfo, claud
 		common.SysError("error unmarshalling stream response: " + err.Error())
 		return service.OpenAIErrorWrapper(err, "stream_response_error", http.StatusInternalServerError)
 	}
-	if claudeResponse.Error.Type != "" {
+	if claudeResponse.Error != nil && claudeResponse.Error.Type != "" {
 		return &dto.OpenAIErrorWithStatusCode{
 			Error: dto.OpenAIError{
 				Code:    "stream_response_error",
@@ -598,7 +600,7 @@ func HandleClaudeResponseData(c *gin.Context, info *relaycommon.RelayInfo, claud
 	if err != nil {
 		return service.OpenAIErrorWrapper(err, "unmarshal_claude_response_failed", http.StatusInternalServerError)
 	}
-	if claudeResponse.Error.Type != "" {
+	if claudeResponse.Error != nil && claudeResponse.Error.Type != "" {
 		return &dto.OpenAIErrorWithStatusCode{
 			Error: dto.OpenAIError{
 				Message: claudeResponse.Error.Message,

relay/channel/cohere/dto.go

@@ -40,8 +40,8 @@ type CohereRerankRequest struct {
 }
 
 type CohereRerankResponseResult struct {
-	Results []dto.RerankResponseDocument `json:"results"`
-	Meta    CohereMeta                   `json:"meta"`
+	Results []dto.RerankResponseResult `json:"results"`
+	Meta    CohereMeta                 `json:"meta"`
 }
 
 type CohereMeta struct {

relay/channel/jina/adaptor.go

@@ -69,7 +69,7 @@ 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 = common_handler.RerankHandler(c, resp)
+		err, usage = common_handler.RerankHandler(c, info, resp)
 	} else if info.RelayMode == constant.RelayModeEmbeddings {
 		err, usage = openai.OpenaiHandler(c, resp, info)
 	}

relay/channel/openai/adaptor.go

@@ -262,7 +262,7 @@ func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, info *relaycom
 	case constant.RelayModeImagesGenerations:
 		err, usage = OpenaiTTSHandler(c, resp, info)
 	case constant.RelayModeRerank:
-		err, usage = common_handler.RerankHandler(c, resp)
+		err, usage = common_handler.RerankHandler(c, info, resp)
 	default:
 		if info.IsStream {
 			err, usage = OaiStreamHandler(c, resp, info)

relay/channel/siliconflow/dto.go

@@ -12,6 +12,6 @@ type SFMeta struct {
 }
 
 type SFRerankResponse struct {
-	Results []dto.RerankResponseDocument `json:"results"`
-	Meta    SFMeta                       `json:"meta"`
+	Results []dto.RerankResponseResult `json:"results"`
+	Meta    SFMeta                     `json:"meta"`
 }

relay/channel/vertex/adaptor.go

@@ -39,8 +39,15 @@ type Adaptor struct {
 }
 
 func (a *Adaptor) ConvertClaudeRequest(c *gin.Context, info *relaycommon.RelayInfo, request *dto.ClaudeRequest) (any, error) {
-	return request, nil
+	if v, ok := claudeModelMap[info.UpstreamModelName]; ok {
+		c.Set("request_model", v)
+	} else {
+		c.Set("request_model", request.Model)
+	}
+	vertexClaudeReq := copyRequest(request, anthropicVersion)
+	return vertexClaudeReq, nil
 }
+
 func (a *Adaptor) ConvertAudioRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.AudioRequest) (io.Reader, error) {
 	//TODO implement me
 	return nil, errors.New("not implemented")

relay/channel/xinference/dto.go

@@ -0,0 +1,11 @@
+package xinference
+
+type XinRerankResponseDocument struct {
+	Document       string  `json:"document,omitempty"`
+	Index          int     `json:"index"`
+	RelevanceScore float64 `json:"relevance_score"`
+}
+
+type XinRerankResponse struct {
+	Results []XinRerankResponseDocument `json:"results"`
+}

relay/common/relay_info.go

@@ -33,6 +33,11 @@ const (
 	RelayFormatClaude = "claude"
 )
 
+type RerankerInfo struct {
+	Documents       []any
+	ReturnDocuments bool
+}
+
 type RelayInfo struct {
 	ChannelType       int
 	ChannelId         int
@@ -71,6 +76,7 @@ type RelayInfo struct {
 	AudioUsage           bool
 	ReasoningEffort      string
 	ChannelSetting       map[string]interface{}
+	ParamOverride        map[string]interface{}
 	UserSetting          map[string]interface{}
 	UserEmail            string
 	UserQuota            int
@@ -78,6 +84,7 @@ type RelayInfo struct {
 	SendResponseCount    int
 	ThinkingContentInfo
 	ClaudeConvertInfo
+	*RerankerInfo
 }
 
 // 定义支持流式选项的通道类型
@@ -111,10 +118,21 @@ func GenRelayInfoClaude(c *gin.Context) *RelayInfo {
 	return info
 }
 
+func GenRelayInfoRerank(c *gin.Context, req *dto.RerankRequest) *RelayInfo {
+	info := GenRelayInfo(c)
+	info.RelayMode = relayconstant.RelayModeRerank
+	info.RerankerInfo = &RerankerInfo{
+		Documents:       req.Documents,
+		ReturnDocuments: req.GetReturnDocuments(),
+	}
+	return info
+}
+
 func GenRelayInfo(c *gin.Context) *RelayInfo {
 	channelType := c.GetInt("channel_type")
 	channelId := c.GetInt("channel_id")
 	channelSetting := c.GetStringMap("channel_setting")
+	paramOverride := c.GetStringMap("param_override")
 
 	tokenId := c.GetInt("token_id")
 	tokenKey := c.GetString("token_key")
@@ -152,6 +170,7 @@ func GenRelayInfo(c *gin.Context) *RelayInfo {
 		ApiKey:         strings.TrimPrefix(c.Request.Header.Get("Authorization"), "Bearer "),
 		Organization:   c.GetString("channel_organization"),
 		ChannelSetting: channelSetting,
+		ParamOverride:  paramOverride,
 		RelayFormat:    RelayFormatOpenAI,
 		ThinkingContentInfo: ThinkingContentInfo{
 			IsFirstThinkingContent:  true,

relay/common_handler/rerank.go

@@ -1,15 +1,17 @@
 package common_handler
 
 import (
-	"encoding/json"
 	"github.com/gin-gonic/gin"
 	"io"
 	"net/http"
+	"one-api/common"
 	"one-api/dto"
+	"one-api/relay/channel/xinference"
+	relaycommon "one-api/relay/common"
 	"one-api/service"
 )
 
-func RerankHandler(c *gin.Context, resp *http.Response) (*dto.OpenAIErrorWithStatusCode, *dto.Usage) {
+func RerankHandler(c *gin.Context, info *relaycommon.RelayInfo, 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
@@ -18,18 +20,49 @@ func RerankHandler(c *gin.Context, resp *http.Response) (*dto.OpenAIErrorWithSta
 	if err != nil {
 		return service.OpenAIErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), nil
 	}
+	if common.DebugEnabled {
+		println("reranker response body: ", string(responseBody))
+	}
 	var jinaResp dto.RerankResponse
-	err = json.Unmarshal(responseBody, &jinaResp)
-	if err != nil {
-		return service.OpenAIErrorWrapper(err, "unmarshal_response_body_failed", http.StatusInternalServerError), nil
+	if info.ChannelType == common.ChannelTypeXinference {
+		var xinRerankResponse xinference.XinRerankResponse
+		err = common.DecodeJson(responseBody, &xinRerankResponse)
+		if err != nil {
+			return service.OpenAIErrorWrapper(err, "unmarshal_response_body_failed", http.StatusInternalServerError), nil
+		}
+		jinaRespResults := make([]dto.RerankResponseResult, len(xinRerankResponse.Results))
+		for i, result := range xinRerankResponse.Results {
+			respResult := dto.RerankResponseResult{
+				Index:          result.Index,
+				RelevanceScore: result.RelevanceScore,
+			}
+			if info.ReturnDocuments {
+				var document any
+				if result.Document == "" {
+					document = info.Documents[result.Index]
+				} else {
+					document = result.Document
+				}
+				respResult.Document = document
+			}
+			jinaRespResults[i] = respResult
+		}
+		jinaResp = dto.RerankResponse{
+			Results: jinaRespResults,
+			Usage: dto.Usage{
+				PromptTokens: info.PromptTokens,
+				TotalTokens:  info.PromptTokens,
+			},
+		}
+	} else {
+		err = common.DecodeJson(responseBody, &jinaResp)
+		if err != nil {
+			return service.OpenAIErrorWrapper(err, "unmarshal_response_body_failed", http.StatusInternalServerError), nil
+		}
+		jinaResp.Usage.PromptTokens = jinaResp.Usage.TotalTokens
 	}
 
-	jsonResponse, err := json.Marshal(jinaResp)
-	if err != nil {
-		return service.OpenAIErrorWrapper(err, "marshal_response_body_failed", http.StatusInternalServerError), nil
-	}
 	c.Writer.Header().Set("Content-Type", "application/json")
-	c.Writer.WriteHeader(resp.StatusCode)
-	_, err = c.Writer.Write(jsonResponse)
+	c.JSON(http.StatusOK, jinaResp)
 	return nil, &jinaResp.Usage
 }

relay/helper/price.go

@@ -20,6 +20,10 @@ type PriceData struct {
 	ShouldPreConsumedQuota int
 }
 
+func (p PriceData) ToSetting() string {
+	return fmt.Sprintf("ModelPrice: %f, ModelRatio: %f, CompletionRatio: %f, CacheRatio: %f, GroupRatio: %f, UsePrice: %t, CacheCreationRatio: %f, ShouldPreConsumedQuota: %d", p.ModelPrice, p.ModelRatio, p.CompletionRatio, p.CacheRatio, p.GroupRatio, p.UsePrice, p.CacheCreationRatio, p.ShouldPreConsumedQuota)
+}
+
 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)
@@ -50,7 +54,8 @@ func ModelPriceHelper(c *gin.Context, info *relaycommon.RelayInfo, promptTokens
 	} else {
 		preConsumedQuota = int(modelPrice * common.QuotaPerUnit * groupRatio)
 	}
-	return PriceData{
+
+	priceData := PriceData{
 		ModelPrice:             modelPrice,
 		ModelRatio:             modelRatio,
 		CompletionRatio:        completionRatio,
@@ -59,5 +64,11 @@ func ModelPriceHelper(c *gin.Context, info *relaycommon.RelayInfo, promptTokens
 		CacheRatio:             cacheRatio,
 		CacheCreationRatio:     cacheCreationRatio,
 		ShouldPreConsumedQuota: preConsumedQuota,
-	}, nil
+	}
+
+	if common.DebugEnabled {
+		println(fmt.Sprintf("model_price_helper result: %s", priceData.ToSetting()))
+	}
+
+	return priceData, nil
 }

relay/relay-text.go

@@ -109,7 +109,7 @@ func TextHelper(c *gin.Context) (openaiErr *dto.OpenAIErrorWithStatusCode) {
 		c.Set("prompt_tokens", promptTokens)
 	}
 
-	priceData, err := helper.ModelPriceHelper(c, relayInfo, promptTokens, int(textRequest.MaxTokens))
+	priceData, err := helper.ModelPriceHelper(c, relayInfo, promptTokens, int(math.Max(float64(textRequest.MaxTokens), float64(textRequest.MaxCompletionTokens))))
 	if err != nil {
 		return service.OpenAIErrorWrapperLocal(err, "model_price_error", http.StatusInternalServerError)
 	}
@@ -168,6 +168,23 @@ func TextHelper(c *gin.Context) (openaiErr *dto.OpenAIErrorWithStatusCode) {
 		if err != nil {
 			return service.OpenAIErrorWrapperLocal(err, "json_marshal_failed", http.StatusInternalServerError)
 		}
+
+		// apply param override
+		if len(relayInfo.ParamOverride) > 0 {
+			reqMap := make(map[string]interface{})
+			err = json.Unmarshal(jsonData, &reqMap)
+			if err != nil {
+				return service.OpenAIErrorWrapperLocal(err, "param_override_unmarshal_failed", http.StatusInternalServerError)
+			}
+			for key, value := range relayInfo.ParamOverride {
+				reqMap[key] = value
+			}
+			jsonData, err = json.Marshal(reqMap)
+			if err != nil {
+				return service.OpenAIErrorWrapperLocal(err, "param_override_marshal_failed", http.StatusInternalServerError)
+			}
+		}
+
 		if common.DebugEnabled {
 			println("requestBody: ", string(jsonData))
 		}
@@ -372,17 +389,18 @@ func postConsumeQuota(ctx *gin.Context, relayInfo *relaycommon.RelayInfo,
 		common.LogError(ctx, fmt.Sprintf("total tokens is 0, cannot consume quota, userId %d, channelId %d, "+
 			"tokenId %d, model %s， pre-consumed quota %d", relayInfo.UserId, relayInfo.ChannelId, relayInfo.TokenId, modelName, preConsumedQuota))
 	} else {
-		quotaDelta := quota - preConsumedQuota
-		if quotaDelta != 0 {
-			err := service.PostConsumeQuota(relayInfo, quotaDelta, preConsumedQuota, true)
-			if err != nil {
-				common.LogError(ctx, "error consuming token remain quota: "+err.Error())
-			}
-		}
 		model.UpdateUserUsedQuotaAndRequestCount(relayInfo.UserId, quota)
 		model.UpdateChannelUsedQuota(relayInfo.ChannelId, quota)
 	}
 
+	quotaDelta := quota - preConsumedQuota
+	if quotaDelta != 0 {
+		err := service.PostConsumeQuota(relayInfo, quotaDelta, preConsumedQuota, true)
+		if err != nil {
+			common.LogError(ctx, "error consuming token remain quota: "+err.Error())
+		}
+	}
+
 	logModel := modelName
 	if strings.HasPrefix(logModel, "gpt-4-gizmo") {
 		logModel = "gpt-4-gizmo-*"

relay/relay_rerank.go

@@ -25,7 +25,6 @@ func getRerankPromptToken(rerankRequest dto.RerankRequest) int {
 }
 
 func RerankHelper(c *gin.Context, relayMode int) (openaiErr *dto.OpenAIErrorWithStatusCode) {
-	relayInfo := relaycommon.GenRelayInfo(c)
 
 	var rerankRequest *dto.RerankRequest
 	err := common.UnmarshalBodyReusable(c, &rerankRequest)
@@ -33,6 +32,9 @@ func RerankHelper(c *gin.Context, relayMode int) (openaiErr *dto.OpenAIErrorWith
 		common.LogError(c, fmt.Sprintf("getAndValidateTextRequest failed: %s", err.Error()))
 		return service.OpenAIErrorWrapperLocal(err, "invalid_text_request", http.StatusBadRequest)
 	}
+
+	relayInfo := relaycommon.GenRelayInfoRerank(c, rerankRequest)
+
 	if rerankRequest.Query == "" {
 		return service.OpenAIErrorWrapperLocal(fmt.Errorf("query is empty"), "invalid_query", http.StatusBadRequest)
 	}

service/quota.go

@@ -243,20 +243,18 @@ func PostClaudeConsumeQuota(ctx *gin.Context, relayInfo *relaycommon.RelayInfo,
 		common.LogError(ctx, fmt.Sprintf("total tokens is 0, cannot consume quota, userId %d, channelId %d, "+
 			"tokenId %d, model %s， pre-consumed quota %d", relayInfo.UserId, relayInfo.ChannelId, relayInfo.TokenId, modelName, preConsumedQuota))
 	} else {
-		//if sensitiveResp != nil {
-		//	logContent += fmt.Sprintf("，敏感词：%s", strings.Join(sensitiveResp.SensitiveWords, ", "))
-		//}
-		quotaDelta := quota - preConsumedQuota
-		if quotaDelta != 0 {
-			err := PostConsumeQuota(relayInfo, quotaDelta, preConsumedQuota, true)
-			if err != nil {
-				common.LogError(ctx, "error consuming token remain quota: "+err.Error())
-			}
-		}
 		model.UpdateUserUsedQuotaAndRequestCount(relayInfo.UserId, quota)
 		model.UpdateChannelUsedQuota(relayInfo.ChannelId, quota)
 	}
 
+	quotaDelta := quota - preConsumedQuota
+	if quotaDelta != 0 {
+		err := PostConsumeQuota(relayInfo, quotaDelta, preConsumedQuota, true)
+		if err != nil {
+			common.LogError(ctx, "error consuming token remain quota: "+err.Error())
+		}
+	}
+
 	other := GenerateClaudeOtherInfo(ctx, relayInfo, modelRatio, groupRatio, completionRatio,
 		cacheTokens, cacheRatio, cacheCreationTokens, cacheCreationRatio, modelPrice)
 	model.RecordConsumeLog(ctx, relayInfo.UserId, relayInfo.ChannelId, promptTokens, completionTokens, modelName,
@@ -318,17 +316,18 @@ func PostAudioConsumeQuota(ctx *gin.Context, relayInfo *relaycommon.RelayInfo,
 		common.LogError(ctx, fmt.Sprintf("total tokens is 0, cannot consume quota, userId %d, channelId %d, "+
 			"tokenId %d, model %s， pre-consumed quota %d", relayInfo.UserId, relayInfo.ChannelId, relayInfo.TokenId, relayInfo.OriginModelName, preConsumedQuota))
 	} else {
-		quotaDelta := quota - preConsumedQuota
-		if quotaDelta != 0 {
-			err := PostConsumeQuota(relayInfo, quotaDelta, preConsumedQuota, true)
-			if err != nil {
-				common.LogError(ctx, "error consuming token remain quota: "+err.Error())
-			}
-		}
 		model.UpdateUserUsedQuotaAndRequestCount(relayInfo.UserId, quota)
 		model.UpdateChannelUsedQuota(relayInfo.ChannelId, quota)
 	}
 
+	quotaDelta := quota - preConsumedQuota
+	if quotaDelta != 0 {
+		err := PostConsumeQuota(relayInfo, quotaDelta, preConsumedQuota, true)
+		if err != nil {
+			common.LogError(ctx, "error consuming token remain quota: "+err.Error())
+		}
+	}
+
 	logModel := relayInfo.OriginModelName
 	if extraContent != "" {
 		logContent += ", " + extraContent

setting/operation_setting/cache_ratio.go

@@ -14,6 +14,8 @@ var defaultCacheRatio = map[string]float64{
 	"o1-preview":                          0.5,
 	"o1-mini-2024-09-12":                  0.5,
 	"o1-mini":                             0.5,
+	"o3-mini":                             0.5,
+	"o3-mini-2025-01-31":                  0.5,
 	"gpt-4o-2024-11-20":                   0.5,
 	"gpt-4o-2024-08-06":                   0.5,
 	"gpt-4o":                              0.5,
@@ -21,6 +23,8 @@ var defaultCacheRatio = map[string]float64{
 	"gpt-4o-mini":                         0.5,
 	"gpt-4o-realtime-preview":             0.5,
 	"gpt-4o-mini-realtime-preview":        0.5,
+	"gpt-4.5-preview":                     0.5,
+	"gpt-4.5-preview-2025-02-27":          0.5,
 	"deepseek-chat":                       0.25,
 	"deepseek-reasoner":                   0.25,
 	"deepseek-coder":                      0.25,

setting/operation_setting/model-ratio.go

@@ -375,6 +375,17 @@ func GetCompletionRatio(name string) float64 {
 			return ratio
 		}
 	}
+	hardCodedRatio, contain := getHardcodedCompletionModelRatio(name)
+	if contain {
+		return hardCodedRatio
+	}
+	if ratio, ok := CompletionRatio[name]; ok {
+		return ratio
+	}
+	return hardCodedRatio
+}
+
+func getHardcodedCompletionModelRatio(name string) (float64, bool) {
 	lowercaseName := strings.ToLower(name)
 	if strings.HasPrefix(name, "gpt-4-gizmo") {
 		name = "gpt-4-gizmo-*"
@@ -385,87 +396,86 @@ func GetCompletionRatio(name string) float64 {
 	if strings.HasPrefix(name, "gpt-4") && !strings.HasSuffix(name, "-all") && !strings.HasSuffix(name, "-gizmo-*") {
 		if strings.HasPrefix(name, "gpt-4o") {
 			if name == "gpt-4o-2024-05-13" {
-				return 3
+				return 3, true
 			}
-			return 4
+			return 4, true
 		}
-		if strings.HasPrefix(name, "gpt-4.5") {
-			return 2
+		// gpt-4.5-preview匹配
+		if strings.HasPrefix(name, "gpt-4.5-preview") {
+			return 2, true
 		}
-		if strings.HasPrefix(name, "gpt-4-turbo") || strings.HasSuffix(name, "preview") {
-			return 3
+		if strings.HasPrefix(name, "gpt-4-turbo") || strings.HasSuffix(name, "gpt-4-1106") || strings.HasSuffix(name, "gpt-4-1105") {
+			return 3, true
 		}
-		return 2
+		// 没有特殊标记的 gpt-4 模型默认倍率为 2
+		return 2, false
 	}
 	if strings.HasPrefix(name, "o1") || strings.HasPrefix(name, "o3") {
-		return 4
+		return 4, true
 	}
 	if name == "chatgpt-4o-latest" {
-		return 3
+		return 3, true
 	}
 	if strings.Contains(name, "claude-instant-1") {
-		return 3
+		return 3, true
 	} else if strings.Contains(name, "claude-2") {
-		return 3
+		return 3, true
 	} else if strings.Contains(name, "claude-3") {
-		return 5
+		return 5, true
 	}
 	if strings.HasPrefix(name, "gpt-3.5") {
 		if name == "gpt-3.5-turbo" || strings.HasSuffix(name, "0125") {
 			// https://openai.com/blog/new-embedding-models-and-api-updates
 			// Updated GPT-3.5 Turbo model and lower pricing
-			return 3
+			return 3, true
 		}
 		if strings.HasSuffix(name, "1106") {
-			return 2
+			return 2, true
 		}
-		return 4.0 / 3.0
+		return 4.0 / 3.0, true
 	}
 	if strings.HasPrefix(name, "mistral-") {
-		return 3
+		return 3, true
 	}
 	if strings.HasPrefix(name, "gemini-") {
-		return 4
+		return 4, true
 	}
 	if strings.HasPrefix(name, "command") {
 		switch name {
 		case "command-r":
-			return 3
+			return 3, true
 		case "command-r-plus":
-			return 5
+			return 5, true
 		case "command-r-08-2024":
-			return 4
+			return 4, true
 		case "command-r-plus-08-2024":
-			return 4
+			return 4, true
 		default:
-			return 4
+			return 4, true
 		}
 	}
 	// hint 只给官方上4倍率，由于开源模型供应商自行定价，不对其进行补全倍率进行强制对齐
 	if lowercaseName == "deepseek-chat" || lowercaseName == "deepseek-reasoner" {
-		return 4
+		return 4, true
 	}
 	if strings.HasPrefix(name, "ERNIE-Speed-") {
-		return 2
+		return 2, true
 	} else if strings.HasPrefix(name, "ERNIE-Lite-") {
-		return 2
+		return 2, true
 	} else if strings.HasPrefix(name, "ERNIE-Character") {
-		return 2
+		return 2, true
 	} else if strings.HasPrefix(name, "ERNIE-Functions") {
-		return 2
+		return 2, true
 	}
 	switch name {
 	case "llama2-70b-4096":
-		return 0.8 / 0.64
+		return 0.8 / 0.64, true
 	case "llama3-8b-8192":
-		return 2
+		return 2, true
 	case "llama3-70b-8192":
-		return 0.79 / 0.59
+		return 0.79 / 0.59, true
 	}
-	if ratio, ok := CompletionRatio[name]; ok {
-		return ratio
-	}
-	return 1
+	return 1, false
 }
 
 func GetAudioRatio(name string) float64 {

web/src/components/LinuxDoIcon.js

@@ -6,17 +6,27 @@ const LinuxDoIcon = (props) => {
     return (
       <svg
         className='icon'
-        viewBox='0 0 24 24'
+        viewBox='0 0 16 16'
         version='1.1'
         xmlns='http://www.w3.org/2000/svg'
         width='1em'
         height='1em'
         {...props}
       >
-        <path
-          d='M19.7,17.6c-0.1-0.2-0.2-0.4-0.2-0.6c0-0.4-0.2-0.7-0.5-1c-0.1-0.1-0.3-0.2-0.4-0.2c0.6-1.8-0.3-3.6-1.3-4.9c0,0,0,0,0,0c-0.8-1.2-2-2.1-1.9-3.7c0-1.9,0.2-5.4-3.3-5.1C8.5,2.3,9.5,6,9.4,7.3c0,1.1-0.5,2.2-1.3,3.1c-0.2,0.2-0.4,0.5-0.5,0.7c-1,1.2-1.5,2.8-1.5,4.3c-0.2,0.2-0.4,0.4-0.5,0.6c-0.1,0.1-0.2,0.2-0.2,0.3c-0.1,0.1-0.3,0.2-0.5,0.3c-0.4,0.1-0.7,0.3-0.9,0.7c-0.1,0.3-0.2,0.7-0.1,1.1c0.1,0.2,0.1,0.4,0,0.7c-0.2,0.4-0.2,0.9,0,1.4c0.3,0.4,0.8,0.5,1.5,0.6c0.5,0,1.1,0.2,1.6,0.4l0,0c0.5,0.3,1.1,0.5,1.7,0.5c0.3,0,0.7-0.1,1-0.2c0.3-0.2,0.5-0.4,0.6-0.7c0.4,0,1-0.2,1.7-0.2c0.6,0,1.2,0.2,2,0.1c0,0.1,0,0.2,0.1,0.3c0.2,0.5,0.7,0.9,1.3,1c0.1,0,0.1,0,0.2,0c0.8-0.1,1.6-0.5,2.1-1.1l0,0c0.4-0.4,0.9-0.7,1.4-0.9c0.6-0.3,1-0.5,1.1-1C20.3,18.6,20.1,18.2,19.7,17.6z M12.8,4.8c0.6,0.1,1.1,0.6,1,1.2c0,0.3-0.1,0.6-0.3,0.9c0,0,0,0-0.1,0c-0.2-0.1-0.3-0.1-0.4-0.2c0.1-0.1,0.1-0.3,0.2-0.5c0-0.4-0.2-0.7-0.4-0.7c-0.3,0-0.5,0.3-0.5,0.7c0,0,0,0.1,0,0.1c-0.1-0.1-0.3-0.1-0.4-0.2c0,0,0-0.1,0-0.1C11.8,5.5,12.2,4.9,12.8,4.8z M12.5,6.8c0.1,0.1,0.3,0.2,0.4,0.2c0.1,0,0.3,0.1,0.4,0.2c0.2,0.1,0.4,0.2,0.4,0.5c0,0.3-0.3,0.6-0.9,0.8c-0.2,0.1-0.3,0.1-0.4,0.2c-0.3,0.2-0.6,0.3-1,0.3c-0.3,0-0.6-0.2-0.8-0.4c-0.1-0.1-0.2-0.2-0.4-0.3C10.1,8.2,9.9,8,9.8,7.7c0-0.1,0.1-0.2,0.2-0.3c0.3-0.2,0.4-0.3,0.5-0.4l0.1-0.1c0.2-0.3,0.6-0.5,1-0.5C11.9,6.5,12.2,6.6,12.5,6.8z M10.4,5c0.4,0,0.7,0.4,0.8,1.1c0,0.1,0,0.1,0,0.2c-0.1,0-0.3,0.1-0.4,0.2c0,0,0-0.1,0-0.2c0-0.3-0.2-0.6-0.4-0.5c-0.2,0-0.3,0.3-0.3,0.6c0,0.2,0.1,0.3,0.2,0.4l0,0c0,0-0.1,0.1-0.2,0.1C9.9,6.7,9.7,6.4,9.7,6.1C9.7,5.5,10,5,10.4,5z M9.4,21.1c-0.7,0.3-1.6,0.2-2.2-0.2c-0.6-0.3-1.1-0.4-1.8-0.4c-0.5-0.1-1-0.1-1.1-0.3c-0.1-0.2-0.1-0.5,0.1-1c0.1-0.3,0.1-0.6,0-0.9c-0.1-0.3-0.1-0.5,0-0.8C4.5,17.2,4.7,17.1,5,17c0.3-0.1,0.5-0.2,0.7-0.4c0.1-0.1,0.2-0.2,0.3-0.4c0.3-0.4,0.5-0.6,0.8-0.6c0.6,0.1,1.1,1,1.5,1.9c0.2,0.3,0.4,0.7,0.7,1c0.4,0.5,0.9,1.2,0.9,1.6C9.9,20.6,9.7,20.9,9.4,21.1z M14.3,18.9c0,0.1,0,0.1-0.1,0.2c-1.2,0.9-2.8,1-4.1,0.3c-0.2-0.3-0.4-0.6-0.6-0.9c0.9-0.1,0.7-1.3-1.2-2.5c-2-1.3-0.6-3.7,0.1-4.8c0.1-0.1,0.1,0-0.3,0.8c-0.3,0.6-0.9,2.1-0.1,3.2c0-0.8,0.2-1.6,0.5-2.4c0.7-1.3,1.2-2.8,1.5-4.3c0.1,0.1,0.1,0.1,0.2,0.1c0.1,0.1,0.2,0.2,0.3,0.2c0.2,0.3,0.6,0.4,0.9,0.4c0,0,0.1,0,0.1,0c0.4,0,0.8-0.1,1.1-0.4c0.1-0.1,0.2-0.2,0.4-0.2c0.3-0.1,0.6-0.3,0.9-0.6c0.4,1.3,0.8,2.5,1.4,3.6c0.4,0.8,0.7,1.6,0.9,2.5c0.3,0,0.7,0.1,1,0.3c0.8,0.4,1.1,0.7,1,1.2c-0.1,0-0.1,0-0.2,0c0-0.3-0.2-0.6-0.9-0.9c-0.7-0.3-1.3-0.3-1.5,0.4c-0.1,0-0.2,0.1-0.3,0.1c-0.8,0.4-0.8,1.5-0.9,2.6C14.5,18.2,14.4,18.5,14.3,18.9z M18.9,19.5c-0.6,0.2-1.1,0.6-1.5,1.1c-0.4,0.6-1.1,1-1.9,0.9c-0.4,0-0.8-0.3-0.9-0.7c-0.1-0.6-0.1-1.2,0.2-1.8c0.1-0.4,0.2-0.7,0.3-1.1c0.1-1.2,0.1-1.9,0.6-2.2h0c0,0.5,0.3,0.8,0.7,1c0.5,0,1-0.1,1.4-0.5c0.1,0,0.1,0,0.2,0c0.3,0,0.5,0,0.7,0.2c0.2,0.2,0.3,0.5,0.3,0.7c0,0.3,0.2,0.6,0.3,0.9c0.5,0.5,0.5,0.8,0.5,0.9C19.7,19.1,19.3,19.3,18.9,19.5z M9.9,7.5c-0.1,0-0.1,0-0.1,0.1c0,0,0,0.1,0.1,0.1c0,0,0,0,0,0c0.1,0,0.1,0.1,0.1,0.1c0.3,0.4,0.8,0.6,1.4,0.7c0.5-0.1,1-0.2,1.5-0.6c0.2-0.1,0.4-0.2,0.6-0.3c0.1,0,0.1-0.1,0.1-0.1c0-0.1,0-0.1-0.1-0.1l0,0c-0.2,0.1-0.5,0.2-0.7,0.3c-0.4,0.3-0.9,0.5-1.4,0.5c-0.5,0-0.9-0.3-1.2-0.6C10.1,7.6,10,7.5,9.9,7.5z'
-          fill='currentColor'
-        />
+        <g id='linuxdo_icon' data-name='linuxdo_icon'>
+          <path
+            d='m7.44,0s.09,0,.13,0c.09,0,.19,0,.28,0,.14,0,.29,0,.43,0,.09,0,.18,0,.27,0q.12,0,.25,0t.26.08c.15.03.29.06.44.08,1.97.38,3.78,1.47,4.95,3.11.04.06.09.12.13.18.67.96,1.15,2.11,1.3,3.28q0,.19.09.26c0,.15,0,.29,0,.44,0,.04,0,.09,0,.13,0,.09,0,.19,0,.28,0,.14,0,.29,0,.43,0,.09,0,.18,0,.27,0,.08,0,.17,0,.25q0,.19-.08.26c-.03.15-.06.29-.08.44-.38,1.97-1.47,3.78-3.11,4.95-.06.04-.12.09-.18.13-.96.67-2.11,1.15-3.28,1.3q-.19,0-.26.09c-.15,0-.29,0-.44,0-.04,0-.09,0-.13,0-.09,0-.19,0-.28,0-.14,0-.29,0-.43,0-.09,0-.18,0-.27,0-.08,0-.17,0-.25,0q-.19,0-.26-.08c-.15-.03-.29-.06-.44-.08-1.97-.38-3.78-1.47-4.95-3.11q-.07-.09-.13-.18c-.67-.96-1.15-2.11-1.3-3.28q0-.19-.09-.26c0-.15,0-.29,0-.44,0-.04,0-.09,0-.13,0-.09,0-.19,0-.28,0-.14,0-.29,0-.43,0-.09,0-.18,0-.27,0-.08,0-.17,0-.25q0-.19.08-.26c.03-.15.06-.29.08-.44.38-1.97,1.47-3.78,3.11-4.95.06-.04.12-.09.18-.13C4.42.73,5.57.26,6.74.1,7,.07,7.15,0,7.44,0Z'
+            fill='#EFEFEF'
+          />
+          <path
+            d='m1.27,11.33h13.45c-.94,1.89-2.51,3.21-4.51,3.88-1.99.59-3.96.37-5.8-.57-1.25-.7-2.67-1.9-3.14-3.3Z'
+            fill='#FEB005'
+          />
+          <path
+            d='m12.54,1.99c.87.7,1.82,1.59,2.18,2.68H1.27c.87-1.74,2.33-3.13,4.2-3.78,2.44-.79,5-.47,7.07,1.1Z'
+            fill='#1D1D1F'
+          />
+        </g>
       </svg>
     );
   }
@@ -24,4 +34,4 @@ const LinuxDoIcon = (props) => {
   return <Icon svg={<CustomIcon />} />;
 };
 
-export default LinuxDoIcon;
+export default LinuxDoIcon;

web/src/i18n/locales/en.json

@@ -1275,6 +1275,7 @@
   "代理站地址": "Base URL",
   "对于官方渠道，new-api已经内置地址，除非是第三方代理站点或者Azure的特殊接入地址，否则不需要填写": "For official channels, the new-api has a built-in address. Unless it is a third-party proxy site or a special Azure access address, there is no need to fill it in",
   "渠道额外设置": "Channel extra settings",
+  "参数覆盖": "Parameters override",
   "模型请求速率限制": "Model request rate limit",
   "启用用户模型请求速率限制（可能会影响高并发性能）": "Enable user model request rate limit (may affect high concurrency performance)",
   "限制周期": "Limit period",

web/src/pages/Channel/EditChannel.js

@@ -983,6 +983,23 @@ const EditChannel = (props) => {
               </Typography.Text>
             </Space>
           </>
+          <>
+            <div style={{ marginTop: 10 }}>
+              <Typography.Text strong>
+                {t('参数覆盖')}：
+              </Typography.Text>
+            </div>
+            <TextArea
+              placeholder={t('此项可选，用于覆盖请求参数。不支持覆盖 stream 参数。为一个 JSON 字符串，例如：') + '\n{\n  "temperature": 0\n}'}
+              name="setting"
+              onChange={(value) => {
+                handleInputChange('param_override', value);
+              }}
+              autosize
+              value={inputs.param_override}
+              autoComplete="new-password"
+            />
+          </>
           {inputs.type === 1 && (
             <>
               <div style={{ marginTop: 10 }}>

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

@@ -27,9 +27,10 @@ export default function GeneralSettings(props) {
   const refForm = useRef();
   const [inputsRow, setInputsRow] = useState(inputs);
 
-  function onChange(value, e) {
-    const name = e.target.id;
-    setInputs((inputs) => ({ ...inputs, [name]: value }));
+  function handleFieldChange(fieldName) {
+    return (value) => {
+      setInputs((inputs) => ({ ...inputs, [fieldName]: value }));
+    };
   }
 
   function onSubmit() {
@@ -98,7 +99,7 @@ export default function GeneralSettings(props) {
                   label={t('充值链接')}
                   initValue={''}
                   placeholder={t('例如发卡网站的购买链接')}
-                  onChange={onChange}
+                  onChange={handleFieldChange('TopUpLink')}
                   showClear
                 />
               </Col>
@@ -108,7 +109,7 @@ export default function GeneralSettings(props) {
                   label={t('文档地址')}
                   initValue={''}
                   placeholder={t('例如 https://docs.newapi.pro')}
-                  onChange={onChange}
+                  onChange={handleFieldChange('general_setting.docs_link')}
                   showClear
                 />
               </Col>
@@ -118,7 +119,7 @@ export default function GeneralSettings(props) {
                   label={t('单位美元额度')}
                   initValue={''}
                   placeholder={t('一单位货币能兑换的额度')}
-                  onChange={onChange}
+                  onChange={handleFieldChange('QuotaPerUnit')}
                   showClear
                   onClick={() => setShowQuotaWarning(true)}
                 />
@@ -129,7 +130,7 @@ export default function GeneralSettings(props) {
                   label={t('失败重试次数')}
                   initValue={''}
                   placeholder={t('失败重试次数')}
-                  onChange={onChange}
+                  onChange={handleFieldChange('RetryTimes')}
                   showClear
                 />
               </Col>
@@ -142,12 +143,7 @@ export default function GeneralSettings(props) {
                   size='default'
                   checkedText='｜'
                   uncheckedText='〇'
-                  onChange={(value) => {
-                    setInputs({
-                      ...inputs,
-                      DisplayInCurrencyEnabled: value,
-                    });
-                  }}
+                  onChange={handleFieldChange('DisplayInCurrencyEnabled')}
                 />
               </Col>
               <Col xs={24} sm={12} md={8} lg={8} xl={8}>
@@ -157,12 +153,7 @@ export default function GeneralSettings(props) {
                   size='default'
                   checkedText='｜'
                   uncheckedText='〇'
-                  onChange={(value) =>
-                    setInputs({
-                      ...inputs,
-                      DisplayTokenStatEnabled: value,
-                    })
-                  }
+                  onChange={handleFieldChange('DisplayTokenStatEnabled')}
                 />
               </Col>
               <Col xs={24} sm={12} md={8} lg={8} xl={8}>
@@ -172,12 +163,7 @@ export default function GeneralSettings(props) {
                   size='default'
                   checkedText='｜'
                   uncheckedText='〇'
-                  onChange={(value) =>
-                    setInputs({
-                      ...inputs,
-                      DefaultCollapseSidebar: value,
-                    })
-                  }
+                  onChange={handleFieldChange('DefaultCollapseSidebar')}
                 />
               </Col>
             </Row>
@@ -189,12 +175,7 @@ export default function GeneralSettings(props) {
                   size='default'
                   checkedText='｜'
                   uncheckedText='〇'
-                  onChange={(value) =>
-                    setInputs({
-                      ...inputs,
-                      DemoSiteEnabled: value
-                    })
-                  }
+                  onChange={handleFieldChange('DemoSiteEnabled')}
                 />
               </Col>
               <Col xs={24} sm={12} md={8} lg={8} xl={8}>
@@ -205,12 +186,7 @@ export default function GeneralSettings(props) {
                   size='default'
                   checkedText='｜'
                   uncheckedText='〇'
-                  onChange={(value) =>
-                    setInputs({
-                      ...inputs,
-                      SelfUseModeEnabled: value
-                    })
-                  }
+                  onChange={handleFieldChange('SelfUseModeEnabled')}
                 />
               </Col>
             </Row>