chuan/CLIProxyAPI
1
0
Fork 0
mirror of https://github.com/router-for-me/CLIProxyAPI.git synced 2026-02-18 04:10:51 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix(executor): use upstream model for thinking config and payload translation

Browse Source
This commit is contained in:
hkfires
2025-12-30 17:49:44 +08:00
parent 857c880f99
commit b055e00c1a
8 changed files with 255 additions and 162 deletions
Download Patch File Download Diff File Expand all files Collapse all files

95
internal/runtime/executor/gemini_cli_executor.go
View File

@@ -75,16 +75,21 @@ func (e *GeminiCLIExecutor) Execute(ctx context.Context, auth *cliproxyauth.Auth
reporter := newUsageReporter(ctx, e.Identifier(), req.Model, auth)
defer reporter.trackFailure(ctx, &err)
upstreamModel := strings.TrimSpace(util.ResolveOriginalModel(req.Model, req.Metadata))
if upstreamModel == "" {
upstreamModel = strings.TrimSpace(req.Model)
}
from := opts.SourceFormat
to := sdktranslator.FromString("gemini-cli")
basePayload := sdktranslator.TranslateRequest(from, to, req.Model, bytes.Clone(req.Payload), false)
basePayload = applyThinkingMetadataCLI(basePayload, req.Metadata, req.Model)
basePayload = util.ApplyGemini3ThinkingLevelFromMetadataCLI(req.Model, req.Metadata, basePayload)
basePayload = util.ApplyDefaultThinkingIfNeededCLI(req.Model, basePayload)
basePayload = util.NormalizeGeminiCLIThinkingBudget(req.Model, basePayload)
basePayload = util.StripThinkingConfigIfUnsupported(req.Model, basePayload)
basePayload = fixGeminiCLIImageAspectRatio(req.Model, basePayload)
basePayload = applyPayloadConfigWithRoot(e.cfg, req.Model, "gemini", "request", basePayload)
basePayload := sdktranslator.TranslateRequest(from, to, upstreamModel, bytes.Clone(req.Payload), false)
basePayload = applyThinkingMetadataCLI(basePayload, req.Metadata, upstreamModel)
basePayload = util.ApplyGemini3ThinkingLevelFromMetadataCLI(upstreamModel, req.Metadata, basePayload)
basePayload = util.ApplyDefaultThinkingIfNeededCLI(upstreamModel, basePayload)
basePayload = util.NormalizeGeminiCLIThinkingBudget(upstreamModel, basePayload)
basePayload = util.StripThinkingConfigIfUnsupported(upstreamModel, basePayload)
basePayload = fixGeminiCLIImageAspectRatio(upstreamModel, basePayload)
basePayload = applyPayloadConfigWithRoot(e.cfg, upstreamModel, "gemini", "request", basePayload)
action := "generateContent"
if req.Metadata != nil {
@@ -94,9 +99,9 @@ func (e *GeminiCLIExecutor) Execute(ctx context.Context, auth *cliproxyauth.Auth
}
projectID := resolveGeminiProjectID(auth)
models := cliPreviewFallbackOrder(req.Model)
if len(models) == 0 || models[0] != req.Model {
models = append([]string{req.Model}, models...)
models := cliPreviewFallbackOrder(upstreamModel)
if len(models) == 0 || models[0] != upstreamModel {
models = append([]string{upstreamModel}, models...)
}
httpClient := newHTTPClient(ctx, e.cfg, auth, 0)
@@ -110,6 +115,10 @@ func (e *GeminiCLIExecutor) Execute(ctx context.Context, auth *cliproxyauth.Auth
var lastStatus int
var lastBody []byte
// NOTE: Model capability checks (thinking config, payload rules, image fixes, etc.) must be
// based on upstreamModel (resolved via oauth-model-mappings). The loop variable attemptModel
// is only used as the concrete model id sent to the upstream Gemini CLI endpoint (and the
// model label passed into response translation) when iterating fallback variants.
for idx, attemptModel := range models {
payload := append([]byte(nil), basePayload...)
if action == "countTokens" {
@@ -214,22 +223,27 @@ func (e *GeminiCLIExecutor) ExecuteStream(ctx context.Context, auth *cliproxyaut
reporter := newUsageReporter(ctx, e.Identifier(), req.Model, auth)
defer reporter.trackFailure(ctx, &err)
upstreamModel := strings.TrimSpace(util.ResolveOriginalModel(req.Model, req.Metadata))
if upstreamModel == "" {
upstreamModel = strings.TrimSpace(req.Model)
}
from := opts.SourceFormat
to := sdktranslator.FromString("gemini-cli")
basePayload := sdktranslator.TranslateRequest(from, to, req.Model, bytes.Clone(req.Payload), true)
basePayload = applyThinkingMetadataCLI(basePayload, req.Metadata, req.Model)
basePayload = util.ApplyGemini3ThinkingLevelFromMetadataCLI(req.Model, req.Metadata, basePayload)
basePayload = util.ApplyDefaultThinkingIfNeededCLI(req.Model, basePayload)
basePayload = util.NormalizeGeminiCLIThinkingBudget(req.Model, basePayload)
basePayload = util.StripThinkingConfigIfUnsupported(req.Model, basePayload)
basePayload = fixGeminiCLIImageAspectRatio(req.Model, basePayload)
basePayload = applyPayloadConfigWithRoot(e.cfg, req.Model, "gemini", "request", basePayload)
basePayload := sdktranslator.TranslateRequest(from, to, upstreamModel, bytes.Clone(req.Payload), true)
basePayload = applyThinkingMetadataCLI(basePayload, req.Metadata, upstreamModel)
basePayload = util.ApplyGemini3ThinkingLevelFromMetadataCLI(upstreamModel, req.Metadata, basePayload)
basePayload = util.ApplyDefaultThinkingIfNeededCLI(upstreamModel, basePayload)
basePayload = util.NormalizeGeminiCLIThinkingBudget(upstreamModel, basePayload)
basePayload = util.StripThinkingConfigIfUnsupported(upstreamModel, basePayload)
basePayload = fixGeminiCLIImageAspectRatio(upstreamModel, basePayload)
basePayload = applyPayloadConfigWithRoot(e.cfg, upstreamModel, "gemini", "request", basePayload)
projectID := resolveGeminiProjectID(auth)
models := cliPreviewFallbackOrder(req.Model)
if len(models) == 0 || models[0] != req.Model {
models = append([]string{req.Model}, models...)
models := cliPreviewFallbackOrder(upstreamModel)
if len(models) == 0 || models[0] != upstreamModel {
models = append([]string{upstreamModel}, models...)
}
httpClient := newHTTPClient(ctx, e.cfg, auth, 0)
@@ -243,6 +257,10 @@ func (e *GeminiCLIExecutor) ExecuteStream(ctx context.Context, auth *cliproxyaut
var lastStatus int
var lastBody []byte
// NOTE: Model capability checks (thinking config, payload rules, image fixes, etc.) must be
// based on upstreamModel (resolved via oauth-model-mappings). The loop variable attemptModel
// is only used as the concrete model id sent to the upstream Gemini CLI endpoint (and the
// model label passed into response translation) when iterating fallback variants.
for idx, attemptModel := range models {
payload := append([]byte(nil), basePayload...)
payload = setJSONField(payload, "project", projectID)
@@ -318,7 +336,7 @@ func (e *GeminiCLIExecutor) ExecuteStream(ctx context.Context, auth *cliproxyaut
out := make(chan cliproxyexecutor.StreamChunk)
stream = out
go func(resp *http.Response, reqBody []byte, attempt string) {
go func(resp *http.Response, reqBody []byte, attemptModel string) {
defer close(out)
defer func() {
if errClose := resp.Body.Close(); errClose != nil {
@@ -336,14 +354,14 @@ func (e *GeminiCLIExecutor) ExecuteStream(ctx context.Context, auth *cliproxyaut
reporter.publish(ctx, detail)
}
if bytes.HasPrefix(line, dataTag) {
segments := sdktranslator.TranslateStream(respCtx, to, from, attempt, bytes.Clone(opts.OriginalRequest), reqBody, bytes.Clone(line), &param)
segments := sdktranslator.TranslateStream(respCtx, to, from, attemptModel, bytes.Clone(opts.OriginalRequest), reqBody, bytes.Clone(line), &param)
for i := range segments {
out <- cliproxyexecutor.StreamChunk{Payload: []byte(segments[i])}
}
}
}
segments := sdktranslator.TranslateStream(respCtx, to, from, attempt, bytes.Clone(opts.OriginalRequest), reqBody, bytes.Clone([]byte("[DONE]")), &param)
segments := sdktranslator.TranslateStream(respCtx, to, from, attemptModel, bytes.Clone(opts.OriginalRequest), reqBody, bytes.Clone([]byte("[DONE]")), &param)
for i := range segments {
out <- cliproxyexecutor.StreamChunk{Payload: []byte(segments[i])}
}
@@ -365,12 +383,12 @@ func (e *GeminiCLIExecutor) ExecuteStream(ctx context.Context, auth *cliproxyaut
appendAPIResponseChunk(ctx, e.cfg, data)
reporter.publish(ctx, parseGeminiCLIUsage(data))
var param any
segments := sdktranslator.TranslateStream(respCtx, to, from, attempt, bytes.Clone(opts.OriginalRequest), reqBody, data, &param)
segments := sdktranslator.TranslateStream(respCtx, to, from, attemptModel, bytes.Clone(opts.OriginalRequest), reqBody, data, &param)
for i := range segments {
out <- cliproxyexecutor.StreamChunk{Payload: []byte(segments[i])}
}
segments = sdktranslator.TranslateStream(respCtx, to, from, attempt, bytes.Clone(opts.OriginalRequest), reqBody, bytes.Clone([]byte("[DONE]")), &param)
segments = sdktranslator.TranslateStream(respCtx, to, from, attemptModel, bytes.Clone(opts.OriginalRequest), reqBody, bytes.Clone([]byte("[DONE]")), &param)
for i := range segments {
out <- cliproxyexecutor.StreamChunk{Payload: []byte(segments[i])}
}
@@ -399,9 +417,14 @@ func (e *GeminiCLIExecutor) CountTokens(ctx context.Context, auth *cliproxyauth.
from := opts.SourceFormat
to := sdktranslator.FromString("gemini-cli")
models := cliPreviewFallbackOrder(req.Model)
if len(models) == 0 || models[0] != req.Model {
models = append([]string{req.Model}, models...)
upstreamModel := strings.TrimSpace(util.ResolveOriginalModel(req.Model, req.Metadata))
if upstreamModel == "" {
upstreamModel = strings.TrimSpace(req.Model)
}
models := cliPreviewFallbackOrder(upstreamModel)
if len(models) == 0 || models[0] != upstreamModel {
models = append([]string{upstreamModel}, models...)
}
httpClient := newHTTPClient(ctx, e.cfg, auth, 0)
@@ -417,15 +440,19 @@ func (e *GeminiCLIExecutor) CountTokens(ctx context.Context, auth *cliproxyauth.
var lastStatus int
var lastBody []byte
// NOTE: Model capability checks (thinking config, payload rules, image fixes, etc.) must be
// based on upstreamModel (resolved via oauth-model-mappings). The loop variable attemptModel
// is only used as the concrete model id sent to the upstream Gemini CLI endpoint when iterating
// fallback variants.
for _, attemptModel := range models {
payload := sdktranslator.TranslateRequest(from, to, attemptModel, bytes.Clone(req.Payload), false)
payload = applyThinkingMetadataCLI(payload, req.Metadata, req.Model)
payload = util.ApplyGemini3ThinkingLevelFromMetadataCLI(req.Model, req.Metadata, payload)
payload = applyThinkingMetadataCLI(payload, req.Metadata, upstreamModel)
payload = util.ApplyGemini3ThinkingLevelFromMetadataCLI(upstreamModel, req.Metadata, payload)
payload = deleteJSONField(payload, "project")
payload = deleteJSONField(payload, "model")
payload = deleteJSONField(payload, "request.safetySettings")
payload = util.StripThinkingConfigIfUnsupported(req.Model, payload)
payload = fixGeminiCLIImageAspectRatio(attemptModel, payload)
payload = util.StripThinkingConfigIfUnsupported(upstreamModel, payload)
payload = fixGeminiCLIImageAspectRatio(upstreamModel, payload)
tok, errTok := tokenSource.Token()
if errTok != nil {