Merge branch 'sub' into feature/subscription

2026-03-30 03:43:39 +00:00 · 2026-02-03 01:29:45 +08:00
parent 3d177f3020 1e2c039f40
commit 11eef1ce77
10 changed files with 150 additions and 58 deletions
--- a/dto/openai_request.go
+++ b/dto/openai_request.go
@@ -817,6 +817,10 @@ type OpenAIResponsesRequest struct {
 	User                 string          `json:"user,omitempty"`
 	MaxToolCalls         uint            `json:"max_tool_calls,omitempty"`
 	Prompt               json.RawMessage `json:"prompt,omitempty"`
+	// qwen
+	EnableThinking json.RawMessage `json:"enable_thinking,omitempty"`
+	// perplexity
+	Preset json.RawMessage `json:"preset,omitempty"`
 }

 func (r *OpenAIResponsesRequest) GetTokenCountMeta() *types.TokenCountMeta {
--- a/relay/channel/ali/adaptor.go
+++ b/relay/channel/ali/adaptor.go
@@ -84,6 +84,8 @@ func (a *Adaptor) GetRequestURL(info *relaycommon.RelayInfo) (string, error) {
 			fullRequestURL = fmt.Sprintf("%s/compatible-mode/v1/embeddings", info.ChannelBaseUrl)
 		case constant.RelayModeRerank:
 			fullRequestURL = fmt.Sprintf("%s/api/v1/services/rerank/text-rerank/text-rerank", info.ChannelBaseUrl)
+		case constant.RelayModeResponses:
+			fullRequestURL = fmt.Sprintf("%s/api/v2/apps/protocols/compatible-mode/v1/responses", info.ChannelBaseUrl)
 		case constant.RelayModeImagesGenerations:
 			if isSyncImageModel(info.OriginModelName) {
 				fullRequestURL = fmt.Sprintf("%s/api/v1/services/aigc/multimodal-generation/generation", info.ChannelBaseUrl)
@@ -210,8 +212,7 @@ func (a *Adaptor) ConvertAudioRequest(c *gin.Context, info *relaycommon.RelayInf
 }

 func (a *Adaptor) ConvertOpenAIResponsesRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.OpenAIResponsesRequest) (any, error) {
-	//TODO implement me
-	return nil, errors.New("not implemented")
+	return request, nil
 }

 func (a *Adaptor) DoRequest(c *gin.Context, info *relaycommon.RelayInfo, requestBody io.Reader) (any, error) {
--- a/relay/channel/api_request.go
+++ b/relay/channel/api_request.go
@@ -98,6 +98,19 @@ func processHeaderOverride(info *common.RelayInfo, c *gin.Context) (map[string]s
 	return headerOverride, nil
 }

+func applyHeaderOverrideToRequest(req *http.Request, headerOverride map[string]string) {
+	if req == nil {
+		return
+	}
+	for key, value := range headerOverride {
+		req.Header.Set(key, value)
+		// set Host in req
+		if strings.EqualFold(key, "Host") {
+			req.Host = value
+		}
+	}
+}
+
 func DoApiRequest(a Adaptor, c *gin.Context, info *common.RelayInfo, requestBody io.Reader) (*http.Response, error) {
 	fullRequestURL, err := a.GetRequestURL(info)
 	if err != nil {
@@ -121,9 +134,7 @@ func DoApiRequest(a Adaptor, c *gin.Context, info *common.RelayInfo, requestBody
 	if err != nil {
 		return nil, err
 	}
-	for key, value := range headerOverride {
-		headers.Set(key, value)
-	}
+	applyHeaderOverrideToRequest(req, headerOverride)
 	resp, err := doRequest(c, req, info)
 	if err != nil {
 		return nil, fmt.Errorf("do request failed: %w", err)
@@ -156,9 +167,7 @@ func DoFormRequest(a Adaptor, c *gin.Context, info *common.RelayInfo, requestBod
 	if err != nil {
 		return nil, err
 	}
-	for key, value := range headerOverride {
-		headers.Set(key, value)
-	}
+	applyHeaderOverrideToRequest(req, headerOverride)
 	resp, err := doRequest(c, req, info)
 	if err != nil {
 		return nil, fmt.Errorf("do request failed: %w", err)
--- a/relay/channel/claude/relay-claude.go
+++ b/relay/channel/claude/relay-claude.go
@@ -437,8 +437,10 @@ func StreamResponseClaude2OpenAI(reqMode int, claudeResponse *dto.ClaudeResponse
 		}
 	} else {
 		if claudeResponse.Type == "message_start" {
-			response.Id = claudeResponse.Message.Id
-			response.Model = claudeResponse.Message.Model
+			if claudeResponse.Message != nil {
+				response.Id = claudeResponse.Message.Id
+				response.Model = claudeResponse.Message.Model
+			}
 			//claudeUsage = &claudeResponse.Message.Usage
 			choice.Delta.SetContentString("")
 			choice.Delta.Role = "assistant"
@@ -589,35 +591,63 @@ type ClaudeResponseInfo struct {
 }

 func FormatClaudeResponseInfo(requestMode int, claudeResponse *dto.ClaudeResponse, oaiResponse *dto.ChatCompletionsStreamResponse, claudeInfo *ClaudeResponseInfo) bool {
+	if claudeInfo == nil {
+		return false
+	}
+	if claudeInfo.Usage == nil {
+		claudeInfo.Usage = &dto.Usage{}
+	}
 	if requestMode == RequestModeCompletion {
 		claudeInfo.ResponseText.WriteString(claudeResponse.Completion)
 	} else {
 		if claudeResponse.Type == "message_start" {
-			claudeInfo.ResponseId = claudeResponse.Message.Id
-			claudeInfo.Model = claudeResponse.Message.Model
+			if claudeResponse.Message != nil {
+				claudeInfo.ResponseId = claudeResponse.Message.Id
+				claudeInfo.Model = claudeResponse.Message.Model
+			}

 			// message_start, 获取usage
-			claudeInfo.Usage.PromptTokens = claudeResponse.Message.Usage.InputTokens
-			claudeInfo.Usage.PromptTokensDetails.CachedTokens = claudeResponse.Message.Usage.CacheReadInputTokens
-			claudeInfo.Usage.PromptTokensDetails.CachedCreationTokens = claudeResponse.Message.Usage.CacheCreationInputTokens
-			claudeInfo.Usage.ClaudeCacheCreation5mTokens = claudeResponse.Message.Usage.GetCacheCreation5mTokens()
-			claudeInfo.Usage.ClaudeCacheCreation1hTokens = claudeResponse.Message.Usage.GetCacheCreation1hTokens()
-			claudeInfo.Usage.CompletionTokens = claudeResponse.Message.Usage.OutputTokens
-		} else if claudeResponse.Type == "content_block_delta" {
-			if claudeResponse.Delta.Text != nil {
-				claudeInfo.ResponseText.WriteString(*claudeResponse.Delta.Text)
+			if claudeResponse.Message != nil && claudeResponse.Message.Usage != nil {
+				claudeInfo.Usage.PromptTokens = claudeResponse.Message.Usage.InputTokens
+				claudeInfo.Usage.PromptTokensDetails.CachedTokens = claudeResponse.Message.Usage.CacheReadInputTokens
+				claudeInfo.Usage.PromptTokensDetails.CachedCreationTokens = claudeResponse.Message.Usage.CacheCreationInputTokens
+				claudeInfo.Usage.ClaudeCacheCreation5mTokens = claudeResponse.Message.Usage.GetCacheCreation5mTokens()
+				claudeInfo.Usage.ClaudeCacheCreation1hTokens = claudeResponse.Message.Usage.GetCacheCreation1hTokens()
+				claudeInfo.Usage.CompletionTokens = claudeResponse.Message.Usage.OutputTokens
 			}
-			if claudeResponse.Delta.Thinking != nil {
-				claudeInfo.ResponseText.WriteString(*claudeResponse.Delta.Thinking)
+		} else if claudeResponse.Type == "content_block_delta" {
+			if claudeResponse.Delta != nil {
+				if claudeResponse.Delta.Text != nil {
+					claudeInfo.ResponseText.WriteString(*claudeResponse.Delta.Text)
+				}
+				if claudeResponse.Delta.Thinking != nil {
+					claudeInfo.ResponseText.WriteString(*claudeResponse.Delta.Thinking)
+				}
 			}
 		} else if claudeResponse.Type == "message_delta" {
 			// 最终的usage获取
-			if claudeResponse.Usage.InputTokens > 0 {
-				// 不叠加，只取最新的
-				claudeInfo.Usage.PromptTokens = claudeResponse.Usage.InputTokens
+			if claudeResponse.Usage != nil {
+				if claudeResponse.Usage.InputTokens > 0 {
+					// 不叠加，只取最新的
+					claudeInfo.Usage.PromptTokens = claudeResponse.Usage.InputTokens
+				}
+				if claudeResponse.Usage.CacheReadInputTokens > 0 {
+					claudeInfo.Usage.PromptTokensDetails.CachedTokens = claudeResponse.Usage.CacheReadInputTokens
+				}
+				if claudeResponse.Usage.CacheCreationInputTokens > 0 {
+					claudeInfo.Usage.PromptTokensDetails.CachedCreationTokens = claudeResponse.Usage.CacheCreationInputTokens
+				}
+				if cacheCreation5m := claudeResponse.Usage.GetCacheCreation5mTokens(); cacheCreation5m > 0 {
+					claudeInfo.Usage.ClaudeCacheCreation5mTokens = cacheCreation5m
+				}
+				if cacheCreation1h := claudeResponse.Usage.GetCacheCreation1hTokens(); cacheCreation1h > 0 {
+					claudeInfo.Usage.ClaudeCacheCreation1hTokens = cacheCreation1h
+				}
+				if claudeResponse.Usage.OutputTokens > 0 {
+					claudeInfo.Usage.CompletionTokens = claudeResponse.Usage.OutputTokens
+				}
+				claudeInfo.Usage.TotalTokens = claudeInfo.Usage.PromptTokens + claudeInfo.Usage.CompletionTokens
 			}
-			claudeInfo.Usage.CompletionTokens = claudeResponse.Usage.OutputTokens
-			claudeInfo.Usage.TotalTokens = claudeInfo.Usage.PromptTokens + claudeInfo.Usage.CompletionTokens

 			// 判断是否完整
 			claudeInfo.Done = true
@@ -657,7 +687,9 @@ func HandleStreamResponseData(c *gin.Context, info *relaycommon.RelayInfo, claud
 		} else {
 			if claudeResponse.Type == "message_start" {
 				// message_start, 获取usage
-				info.UpstreamModelName = claudeResponse.Message.Model
+				if claudeResponse.Message != nil {
+					info.UpstreamModelName = claudeResponse.Message.Model
+				}
 			} else if claudeResponse.Type == "content_block_delta" {
 			} else if claudeResponse.Type == "message_delta" {
 			}
@@ -745,13 +777,18 @@ func HandleClaudeResponseData(c *gin.Context, info *relaycommon.RelayInfo, claud
 	if requestMode == RequestModeCompletion {
 		claudeInfo.Usage = service.ResponseText2Usage(c, claudeResponse.Completion, info.UpstreamModelName, info.GetEstimatePromptTokens())
 	} else {
-		claudeInfo.Usage.PromptTokens = claudeResponse.Usage.InputTokens
-		claudeInfo.Usage.CompletionTokens = claudeResponse.Usage.OutputTokens
-		claudeInfo.Usage.TotalTokens = claudeResponse.Usage.InputTokens + claudeResponse.Usage.OutputTokens
-		claudeInfo.Usage.PromptTokensDetails.CachedTokens = claudeResponse.Usage.CacheReadInputTokens
-		claudeInfo.Usage.PromptTokensDetails.CachedCreationTokens = claudeResponse.Usage.CacheCreationInputTokens
-		claudeInfo.Usage.ClaudeCacheCreation5mTokens = claudeResponse.Usage.GetCacheCreation5mTokens()
-		claudeInfo.Usage.ClaudeCacheCreation1hTokens = claudeResponse.Usage.GetCacheCreation1hTokens()
+		if claudeInfo.Usage == nil {
+			claudeInfo.Usage = &dto.Usage{}
+		}
+		if claudeResponse.Usage != nil {
+			claudeInfo.Usage.PromptTokens = claudeResponse.Usage.InputTokens
+			claudeInfo.Usage.CompletionTokens = claudeResponse.Usage.OutputTokens
+			claudeInfo.Usage.TotalTokens = claudeResponse.Usage.InputTokens + claudeResponse.Usage.OutputTokens
+			claudeInfo.Usage.PromptTokensDetails.CachedTokens = claudeResponse.Usage.CacheReadInputTokens
+			claudeInfo.Usage.PromptTokensDetails.CachedCreationTokens = claudeResponse.Usage.CacheCreationInputTokens
+			claudeInfo.Usage.ClaudeCacheCreation5mTokens = claudeResponse.Usage.GetCacheCreation5mTokens()
+			claudeInfo.Usage.ClaudeCacheCreation1hTokens = claudeResponse.Usage.GetCacheCreation1hTokens()
+		}
 	}
 	var responseData []byte
 	switch info.RelayFormat {
@@ -766,7 +803,7 @@ func HandleClaudeResponseData(c *gin.Context, info *relaycommon.RelayInfo, claud
 		responseData = data
 	}

-	if claudeResponse.Usage.ServerToolUse != nil && claudeResponse.Usage.ServerToolUse.WebSearchRequests > 0 {
+	if claudeResponse.Usage != nil && claudeResponse.Usage.ServerToolUse != nil && claudeResponse.Usage.ServerToolUse.WebSearchRequests > 0 {
 		c.Set("claude_web_search_requests", claudeResponse.Usage.ServerToolUse.WebSearchRequests)
 	}

--- a/relay/channel/perplexity/adaptor.go
+++ b/relay/channel/perplexity/adaptor.go
@@ -10,6 +10,7 @@ import (
 	"github.com/QuantumNous/new-api/relay/channel"
 	"github.com/QuantumNous/new-api/relay/channel/openai"
 	relaycommon "github.com/QuantumNous/new-api/relay/common"
+	relayconstant "github.com/QuantumNous/new-api/relay/constant"
 	"github.com/QuantumNous/new-api/types"

 	"github.com/gin-gonic/gin"
@@ -42,6 +43,9 @@ func (a *Adaptor) Init(info *relaycommon.RelayInfo) {
 }

 func (a *Adaptor) GetRequestURL(info *relaycommon.RelayInfo) (string, error) {
+	if info.RelayMode == relayconstant.RelayModeResponses {
+		return fmt.Sprintf("%s/v1/responses", info.ChannelBaseUrl), nil
+	}
 	return fmt.Sprintf("%s/chat/completions", info.ChannelBaseUrl), nil
 }

@@ -71,8 +75,7 @@ func (a *Adaptor) ConvertEmbeddingRequest(c *gin.Context, info *relaycommon.Rela
 }

 func (a *Adaptor) ConvertOpenAIResponsesRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.OpenAIResponsesRequest) (any, error) {
-	// TODO implement me
-	return nil, errors.New("not implemented")
+	return request, nil
 }

 func (a *Adaptor) DoRequest(c *gin.Context, info *relaycommon.RelayInfo, requestBody io.Reader) (any, error) {
--- a/relay/channel/vertex/adaptor.go
+++ b/relay/channel/vertex/adaptor.go
@@ -24,9 +24,9 @@ import (
 )

 const (
-	RequestModeClaude = 1
-	RequestModeGemini = 2
-	RequestModeLlama  = 3
+	RequestModeClaude     = 1
+	RequestModeGemini     = 2
+	RequestModeOpenSource = 3
 )

 var claudeModelMap = map[string]string{
@@ -115,7 +115,7 @@ func (a *Adaptor) Init(info *relaycommon.RelayInfo) {
 	} else if strings.Contains(info.UpstreamModelName, "llama") ||
 		// open source models
 		strings.Contains(info.UpstreamModelName, "-maas") {
-		a.RequestMode = RequestModeLlama
+		a.RequestMode = RequestModeOpenSource
 	} else {
 		a.RequestMode = RequestModeGemini
 	}
@@ -166,10 +166,9 @@ func (a *Adaptor) getRequestUrl(info *relaycommon.RelayInfo, modelName, suffix s
 					suffix,
 				), nil
 			}
-		} else if a.RequestMode == RequestModeLlama {
+		} else if a.RequestMode == RequestModeOpenSource {
 			return fmt.Sprintf(
-				"https://%s-aiplatform.googleapis.com/v1beta1/projects/%s/locations/%s/endpoints/openapi/chat/completions",
-				region,
+				"https://aiplatform.googleapis.com/v1beta1/projects/%s/locations/%s/endpoints/openapi/chat/completions",
 				adc.ProjectID,
 				region,
 			), nil
@@ -242,7 +241,7 @@ func (a *Adaptor) GetRequestURL(info *relaycommon.RelayInfo) (string, error) {
 			model = v
 		}
 		return a.getRequestUrl(info, model, suffix)
-	} else if a.RequestMode == RequestModeLlama {
+	} else if a.RequestMode == RequestModeOpenSource {
 		return a.getRequestUrl(info, "", "")
 	}
 	return "", errors.New("unsupported request mode")
@@ -340,7 +339,7 @@ func (a *Adaptor) ConvertOpenAIRequest(c *gin.Context, info *relaycommon.RelayIn
 		}
 		c.Set("request_model", request.Model)
 		return geminiRequest, nil
-	} else if a.RequestMode == RequestModeLlama {
+	} else if a.RequestMode == RequestModeOpenSource {
 		return request, nil
 	}
 	return nil, errors.New("unsupported request mode")
@@ -375,7 +374,7 @@ func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, info *relaycom
 			} else {
 				return gemini.GeminiChatStreamHandler(c, info, resp)
 			}
-		case RequestModeLlama:
+		case RequestModeOpenSource:
 			return openai.OaiStreamHandler(c, info, resp)
 		}
 	} else {
@@ -391,7 +390,7 @@ func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, info *relaycom
 				}
 				return gemini.GeminiChatHandler(c, info, resp)
 			}
-		case RequestModeLlama:
+		case RequestModeOpenSource:
 			return openai.OpenaiHandler(c, info, resp)
 		}
 	}
--- a/service/file_decoder.go
+++ b/service/file_decoder.go
@@ -231,6 +231,8 @@ func GetMimeTypeByExtension(ext string) string {
 		return "image/png"
 	case "gif":
 		return "image/gif"
+	case "jfif":
+		return "image/jpeg"

 	// Audio files
 	case "mp3":
--- a/setting/operation_setting/channel_affinity_setting.go
+++ b/setting/operation_setting/channel_affinity_setting.go
@@ -33,11 +33,40 @@ type ChannelAffinitySetting struct {
 }

 var channelAffinitySetting = ChannelAffinitySetting{
-	Enabled:           false,
+	Enabled:           true,
 	SwitchOnSuccess:   true,
 	MaxEntries:        100_000,
 	DefaultTTLSeconds: 3600,
-	Rules:             []ChannelAffinityRule{},
+	Rules: []ChannelAffinityRule{
+		{
+			Name:       "codex trace",
+			ModelRegex: []string{"^gpt-.*$"},
+			PathRegex:  []string{"/v1/responses"},
+			KeySources: []ChannelAffinityKeySource{
+				{Type: "gjson", Path: "prompt_cache_key"},
+			},
+			ValueRegex:         "",
+			TTLSeconds:         0,
+			SkipRetryOnFailure: false,
+			IncludeUsingGroup:  true,
+			IncludeRuleName:    true,
+			UserAgentInclude:   nil,
+		},
+		{
+			Name:       "claude code trace",
+			ModelRegex: []string{"^claude-.*$"},
+			PathRegex:  []string{"/v1/messages"},
+			KeySources: []ChannelAffinityKeySource{
+				{Type: "gjson", Path: "metadata.user_id"},
+			},
+			ValueRegex:         "",
+			TTLSeconds:         0,
+			SkipRetryOnFailure: false,
+			IncludeUsingGroup:  true,
+			IncludeRuleName:    true,
+			UserAgentInclude:   nil,
+		},
+	},
 }

 func init() {
--- a/web/src/components/table/task-logs/modals/ContentModal.jsx
+++ b/web/src/components/table/task-logs/modals/ContentModal.jsx
@@ -120,7 +120,7 @@ const ContentModal = ({
    }

    return (
-      <div style={{ position: 'relative' }}>
+      <div style={{ position: 'relative', height: '100%' }}>
        {isLoading && (
          <div
            style={{
@@ -137,7 +137,13 @@ const ContentModal = ({
        <video
          src={modalContent}
          controls
-          style={{ width: '100%' }}
+          style={{
+            width: '100%',
+            height: '100%',
+            maxWidth: '100%',
+            maxHeight: '100%',
+            objectFit: 'contain',
+          }}
          autoPlay
          crossOrigin='anonymous'
          onError={handleVideoError}
@@ -155,11 +161,13 @@ const ContentModal = ({
      onCancel={() => setIsModalOpen(false)}
      closable={null}
      bodyStyle={{
-        height: isVideo ? '450px' : '400px',
+        height: isVideo ? '70vh' : '400px',
+        maxHeight: '80vh',
        overflow: 'auto',
        padding: isVideo && videoError ? '0' : '24px',
      }}
-      width={800}
+      width={isVideo ? '90vw' : 800}
+      style={isVideo ? { maxWidth: 960 } : undefined}
    >
      {isVideo ? (
        renderVideoContent()
--- a/web/src/pages/Setting/Operation/SettingsChannelAffinity.jsx
+++ b/web/src/pages/Setting/Operation/SettingsChannelAffinity.jsx
@@ -67,7 +67,7 @@ const KEY_SOURCE_TYPES = [

 const RULE_TEMPLATES = {
  codex: {
-    name: 'codex优选',
+    name: 'codex trace',
    model_regex: ['^gpt-.*$'],
    path_regex: ['/v1/responses'],
    key_sources: [{ type: 'gjson', path: 'prompt_cache_key' }],
@@ -78,7 +78,7 @@ const RULE_TEMPLATES = {
    include_rule_name: true,
  },
  claudeCode: {
-    name: 'claude-code优选',
+    name: 'claude-code trace',
    model_regex: ['^claude-.*$'],
    path_regex: ['/v1/messages'],
    key_sources: [{ type: 'gjson', path: 'metadata.user_id' }],