Merge pull request #1 from router-for-me/Plus

v6.5.31
Merge branch 'main' into Plus
2026-03-17 14:33:06 +00:00 · 2025-12-01 09:10:55 +08:00 · 2025-12-01 09:10:29 +08:00 · 2025-12-01 09:00:27 +08:00 · 2025-12-01 07:22:42 +08:00 · 2025-11-30 15:20:15 -08:00
7 changed files with 54 additions and 52 deletions
--- a/internal/registry/model_definitions.go
+++ b/internal/registry/model_definitions.go
@@ -222,6 +222,7 @@ func GetGeminiModels() []*ModelInfo {
 			InputTokenLimit:            1048576,
 			OutputTokenLimit:           65536,
 			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
 		},
 	}
 }
@@ -301,6 +302,7 @@ func GetGeminiVertexModels() []*ModelInfo {
 			InputTokenLimit:            1048576,
 			OutputTokenLimit:           65536,
 			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
 		},
 	}
 }
--- a/internal/runtime/executor/aistudio_executor.go
+++ b/internal/runtime/executor/aistudio_executor.go
@@ -308,13 +308,7 @@ func (e *AIStudioExecutor) translateRequest(req cliproxyexecutor.Request, opts c
 	from := opts.SourceFormat
 	to := sdktranslator.FromString("gemini")
 	payload := sdktranslator.TranslateRequest(from, to, req.Model, bytes.Clone(req.Payload), stream)
-	if budgetOverride, includeOverride, ok := util.GeminiThinkingFromMetadata(req.Metadata); ok && util.ModelSupportsThinking(req.Model) {
-		if budgetOverride != nil {
-			norm := util.NormalizeThinkingBudget(req.Model, *budgetOverride)
-			budgetOverride = &norm
-		}
-		payload = util.ApplyGeminiThinkingConfig(payload, budgetOverride, includeOverride)
-	}
+	payload = applyThinkingMetadata(payload, req.Metadata, req.Model)
 	payload = util.ConvertThinkingLevelToBudget(payload)
 	payload = util.StripThinkingConfigIfUnsupported(req.Model, payload)
 	payload = fixGeminiImageAspectRatio(req.Model, payload)
--- a/internal/runtime/executor/antigravity_executor.go
+++ b/internal/runtime/executor/antigravity_executor.go
@@ -75,6 +75,8 @@ func (e *AntigravityExecutor) Execute(ctx context.Context, auth *cliproxyauth.Au
 	to := sdktranslator.FromString("antigravity")
 	translated := sdktranslator.TranslateRequest(from, to, req.Model, bytes.Clone(req.Payload), false)

+	translated = applyThinkingMetadataCLI(translated, req.Metadata, req.Model)
+
 	baseURLs := antigravityBaseURLFallbackOrder(auth)
 	httpClient := newProxyAwareHTTPClient(ctx, e.cfg, auth, 0)

@@ -166,6 +168,8 @@ func (e *AntigravityExecutor) ExecuteStream(ctx context.Context, auth *cliproxya
 	to := sdktranslator.FromString("antigravity")
 	translated := sdktranslator.TranslateRequest(from, to, req.Model, bytes.Clone(req.Payload), true)

+	translated = applyThinkingMetadataCLI(translated, req.Metadata, req.Model)
+
 	baseURLs := antigravityBaseURLFallbackOrder(auth)
 	httpClient := newProxyAwareHTTPClient(ctx, e.cfg, auth, 0)

--- a/internal/runtime/executor/gemini_cli_executor.go
+++ b/internal/runtime/executor/gemini_cli_executor.go
@@ -62,15 +62,8 @@ func (e *GeminiCLIExecutor) Execute(ctx context.Context, auth *cliproxyauth.Auth

 	from := opts.SourceFormat
 	to := sdktranslator.FromString("gemini-cli")
-	budgetOverride, includeOverride, hasOverride := util.GeminiThinkingFromMetadata(req.Metadata)
 	basePayload := sdktranslator.TranslateRequest(from, to, req.Model, bytes.Clone(req.Payload), false)
-	if hasOverride && util.ModelSupportsThinking(req.Model) {
-		if budgetOverride != nil {
-			norm := util.NormalizeThinkingBudget(req.Model, *budgetOverride)
-			budgetOverride = &norm
-		}
-		basePayload = util.ApplyGeminiCLIThinkingConfig(basePayload, budgetOverride, includeOverride)
-	}
+	basePayload = applyThinkingMetadataCLI(basePayload, req.Metadata, req.Model)
 	basePayload = util.StripThinkingConfigIfUnsupported(req.Model, basePayload)
 	basePayload = fixGeminiCLIImageAspectRatio(req.Model, basePayload)
 	basePayload = applyPayloadConfigWithRoot(e.cfg, req.Model, "gemini", "request", basePayload)
@@ -204,15 +197,8 @@ func (e *GeminiCLIExecutor) ExecuteStream(ctx context.Context, auth *cliproxyaut

 	from := opts.SourceFormat
 	to := sdktranslator.FromString("gemini-cli")
-	budgetOverride, includeOverride, hasOverride := util.GeminiThinkingFromMetadata(req.Metadata)
 	basePayload := sdktranslator.TranslateRequest(from, to, req.Model, bytes.Clone(req.Payload), true)
-	if hasOverride && util.ModelSupportsThinking(req.Model) {
-		if budgetOverride != nil {
-			norm := util.NormalizeThinkingBudget(req.Model, *budgetOverride)
-			budgetOverride = &norm
-		}
-		basePayload = util.ApplyGeminiCLIThinkingConfig(basePayload, budgetOverride, includeOverride)
-	}
+	basePayload = applyThinkingMetadataCLI(basePayload, req.Metadata, req.Model)
 	basePayload = util.StripThinkingConfigIfUnsupported(req.Model, basePayload)
 	basePayload = fixGeminiCLIImageAspectRatio(req.Model, basePayload)
 	basePayload = applyPayloadConfigWithRoot(e.cfg, req.Model, "gemini", "request", basePayload)
@@ -408,16 +394,9 @@ func (e *GeminiCLIExecutor) CountTokens(ctx context.Context, auth *cliproxyauth.
 	var lastStatus int
 	var lastBody []byte

-	budgetOverride, includeOverride, hasOverride := util.GeminiThinkingFromMetadata(req.Metadata)
 	for _, attemptModel := range models {
 		payload := sdktranslator.TranslateRequest(from, to, attemptModel, bytes.Clone(req.Payload), false)
-		if hasOverride && util.ModelSupportsThinking(req.Model) {
-			if budgetOverride != nil {
-				norm := util.NormalizeThinkingBudget(req.Model, *budgetOverride)
-				budgetOverride = &norm
-			}
-			payload = util.ApplyGeminiCLIThinkingConfig(payload, budgetOverride, includeOverride)
-		}
+		payload = applyThinkingMetadataCLI(payload, req.Metadata, req.Model)
 		payload = deleteJSONField(payload, "project")
 		payload = deleteJSONField(payload, "model")
 		payload = deleteJSONField(payload, "request.safetySettings")
--- a/internal/runtime/executor/gemini_executor.go
+++ b/internal/runtime/executor/gemini_executor.go
@@ -79,13 +79,7 @@ func (e *GeminiExecutor) Execute(ctx context.Context, auth *cliproxyauth.Auth, r
 	from := opts.SourceFormat
 	to := sdktranslator.FromString("gemini")
 	body := sdktranslator.TranslateRequest(from, to, req.Model, bytes.Clone(req.Payload), false)
-	if budgetOverride, includeOverride, ok := util.GeminiThinkingFromMetadata(req.Metadata); ok && util.ModelSupportsThinking(req.Model) {
-		if budgetOverride != nil {
-			norm := util.NormalizeThinkingBudget(req.Model, *budgetOverride)
-			budgetOverride = &norm
-		}
-		body = util.ApplyGeminiThinkingConfig(body, budgetOverride, includeOverride)
-	}
+	body = applyThinkingMetadata(body, req.Metadata, req.Model)
 	body = util.StripThinkingConfigIfUnsupported(req.Model, body)
 	body = fixGeminiImageAspectRatio(req.Model, body)
 	body = applyPayloadConfig(e.cfg, req.Model, body)
@@ -174,13 +168,7 @@ func (e *GeminiExecutor) ExecuteStream(ctx context.Context, auth *cliproxyauth.A
 	from := opts.SourceFormat
 	to := sdktranslator.FromString("gemini")
 	body := sdktranslator.TranslateRequest(from, to, req.Model, bytes.Clone(req.Payload), true)
-	if budgetOverride, includeOverride, ok := util.GeminiThinkingFromMetadata(req.Metadata); ok && util.ModelSupportsThinking(req.Model) {
-		if budgetOverride != nil {
-			norm := util.NormalizeThinkingBudget(req.Model, *budgetOverride)
-			budgetOverride = &norm
-		}
-		body = util.ApplyGeminiThinkingConfig(body, budgetOverride, includeOverride)
-	}
+	body = applyThinkingMetadata(body, req.Metadata, req.Model)
 	body = util.StripThinkingConfigIfUnsupported(req.Model, body)
 	body = fixGeminiImageAspectRatio(req.Model, body)
 	body = applyPayloadConfig(e.cfg, req.Model, body)
@@ -288,13 +276,7 @@ func (e *GeminiExecutor) CountTokens(ctx context.Context, auth *cliproxyauth.Aut
 	from := opts.SourceFormat
 	to := sdktranslator.FromString("gemini")
 	translatedReq := sdktranslator.TranslateRequest(from, to, req.Model, bytes.Clone(req.Payload), false)
-	if budgetOverride, includeOverride, ok := util.GeminiThinkingFromMetadata(req.Metadata); ok && util.ModelSupportsThinking(req.Model) {
-		if budgetOverride != nil {
-			norm := util.NormalizeThinkingBudget(req.Model, *budgetOverride)
-			budgetOverride = &norm
-		}
-		translatedReq = util.ApplyGeminiThinkingConfig(translatedReq, budgetOverride, includeOverride)
-	}
+	translatedReq = applyThinkingMetadata(translatedReq, req.Metadata, req.Model)
 	translatedReq = util.StripThinkingConfigIfUnsupported(req.Model, translatedReq)
 	translatedReq = fixGeminiImageAspectRatio(req.Model, translatedReq)
 	respCtx := context.WithValue(ctx, "alt", opts.Alt)
--- a/internal/runtime/executor/payload_helpers.go
+++ b/internal/runtime/executor/payload_helpers.go
@@ -4,10 +4,42 @@ import (
 	"strings"

 	"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
+	"github.com/router-for-me/CLIProxyAPI/v6/internal/util"
 	"github.com/tidwall/gjson"
 	"github.com/tidwall/sjson"
 )

+// applyThinkingMetadata applies thinking config from model suffix metadata (e.g., -reasoning, -thinking-N)
+// for standard Gemini format payloads. It normalizes the budget when the model supports thinking.
+func applyThinkingMetadata(payload []byte, metadata map[string]any, model string) []byte {
+	budgetOverride, includeOverride, ok := util.GeminiThinkingFromMetadata(metadata)
+	if !ok {
+		return payload
+	}
+	if !util.ModelSupportsThinking(model) {
+		return payload
+	}
+	if budgetOverride != nil {
+		norm := util.NormalizeThinkingBudget(model, *budgetOverride)
+		budgetOverride = &norm
+	}
+	return util.ApplyGeminiThinkingConfig(payload, budgetOverride, includeOverride)
+}
+
+// applyThinkingMetadataCLI applies thinking config from model suffix metadata (e.g., -reasoning, -thinking-N)
+// for Gemini CLI format payloads (nested under "request"). It normalizes the budget when the model supports thinking.
+func applyThinkingMetadataCLI(payload []byte, metadata map[string]any, model string) []byte {
+	budgetOverride, includeOverride, ok := util.GeminiThinkingFromMetadata(metadata)
+	if !ok {
+		return payload
+	}
+	if budgetOverride != nil && util.ModelSupportsThinking(model) {
+		norm := util.NormalizeThinkingBudget(model, *budgetOverride)
+		budgetOverride = &norm
+	}
+	return util.ApplyGeminiCLIThinkingConfig(payload, budgetOverride, includeOverride)
+}
+
 // applyPayloadConfig applies payload default and override rules from configuration
 // to the given JSON payload for the specified model.
 // Defaults only fill missing fields, while overrides always overwrite existing values.
--- a/internal/util/gemini_thinking.go
+++ b/internal/util/gemini_thinking.go
@@ -34,6 +34,15 @@ func ParseGeminiThinkingSuffix(model string) (string, *int, *bool, bool) {
 		return base, &budgetValue, &include, true
 	}

+	// Handle "-reasoning" suffix: enables thinking with dynamic budget (-1)
+	// Maps: gemini-2.5-flash-reasoning -> gemini-2.5-flash with thinkingBudget=-1
+	if strings.HasSuffix(lower, "-reasoning") {
+		base := model[:len(model)-len("-reasoning")]
+		budgetValue := -1 // Dynamic budget
+		include := true
+		return base, &budgetValue, &include, true
+	}
+
 	idx := strings.LastIndex(lower, "-thinking-")
 	if idx == -1 {
 		return model, nil, nil, false
Author	SHA1	Message	Date
Luis Pater	c6ecdf4a7e	Merge pull request #1 from router-for-me/Plus v6.5.31	2025-12-01 09:10:55 +08:00
Luis Pater	838f47a7e9	Merge branch 'main' into Plus	2025-12-01 09:10:29 +08:00
Luis Pater	54e24110ec	Merge pull request #386 from auroraflux/feat/dedupe-thinking-metadata-helpers refactor(executor): dedupe thinking metadata helpers across Gemini executors	2025-12-01 09:00:27 +08:00
Luis Pater	717c703bff	docs(readme): add CCS (Claude Code Switch) to projects list	2025-12-01 07:22:42 +08:00
auroraflux	1c6f4be8ae	refactor(executor): dedupe thinking metadata helpers across Gemini executors Extract applyThinkingMetadata and applyThinkingMetadataCLI helpers to payload_helpers.go and use them across all four Gemini-based executors: - gemini_executor.go (Execute, ExecuteStream, CountTokens) - gemini_cli_executor.go (Execute, ExecuteStream, CountTokens) - aistudio_executor.go (translateRequest) - antigravity_executor.go (Execute, ExecuteStream) This eliminates code duplication introduced in the -reasoning suffix PR and centralizes the thinking config application logic. Net reduction: 28 lines of code.	2025-11-30 15:20:15 -08:00
Luis Pater	0de2560cee	Merge pull request #379 from kaitranntt/docs/add-ccs-project docs: add CCS (Claude Code Switch) to projects list	2025-12-01 07:20:04 +08:00
Kai (Tam Nhu) Tran	85eb926482	fix: change AGY to Antigravity	2025-11-30 12:43:12 -05:00
Kai (Tam Nhu) Tran	c52ef08e67	docs: add CCS to projects list	2025-11-30 12:40:35 -05:00
Luis Pater	5b01eba943	Merge branch 'router-for-me:main' into main	2025-11-30 21:30:49 +08:00
Luis Pater	cb580cd083	Merge pull request #377 from router-for-me/gemini feat(registry): add thinking support to gemini models	2025-11-30 21:27:54 +08:00
hkfires	75e278c7a5	feat(registry): add thinking support to gemini models	2025-11-30 20:56:29 +08:00
Luis Pater	73208c4e55	Merge pull request #376 from auroraflux/feat/reasoning-suffix-support feat(util): add -reasoning suffix support for Gemini models	2025-11-30 20:55:38 +08:00
auroraflux	32d3809f8c	feat(util): add -reasoning suffix support for Gemini models Adds support for the `-reasoning` model name suffix which enables thinking/reasoning mode with dynamic budget. This allows clients to request reasoning-enabled inference using model names like `gemini-2.5-flash-reasoning` without explicit configuration. The suffix is normalized to the base model (e.g., gemini-2.5-flash) with thinkingBudget=-1 (dynamic) and include_thoughts=true. Follows the existing pattern established by -nothinking and -thinking-N suffixes.	2025-11-30 01:18:57 -08:00