chuan/CLIProxyAPI
1
0
Fork 0
mirror of https://github.com/router-for-me/CLIProxyAPI.git synced 2026-02-03 13:00:52 +08:00
Code Issues Packages Projects Releases Wiki Activity

feat(util): implement dynamic thinking suffix normalization and refactor budget resolution logic

Browse Source 
- Added support for parsing and normalizing dynamic thinking model suffixes.
- Centralized budget resolution across executors and payload helpers.
- Retired legacy Gemini-specific thinking handlers in favor of unified logic.
- Updated executors to use metadata-based thinking configuration.
- Added `ResolveOriginalModel` utility for resolving normalized upstream models using request metadata.
- Updated executors (Gemini, Codex, iFlow, OpenAI, Qwen) to incorporate upstream model resolution and substitute model values in payloads and request URLs.
- Ensured fallbacks handle cases with missing or malformed metadata to derive models robustly.
- Refactored upstream model resolution to dynamically incorporate metadata for selecting and normalizing models.
- Improved handling of thinking configurations and model overrides in executors.
- Removed hardcoded thinking model entries and migrated logic to metadata-based resolution.
- Updated payload mutations to always include the resolved model.
This commit is contained in:
Luis Pater
2025-12-11 01:16:56 +08:00
parent e717939edb
commit 423ce97665
13 changed files with 579 additions and 647 deletions
Download Patch File Download Diff File Expand all files Collapse all files

96
internal/runtime/executor/codex_executor.go
View File

@@ -49,14 +49,14 @@ func (e *CodexExecutor) Execute(ctx context.Context, auth *cliproxyauth.Auth, re
reporter := newUsageReporter(ctx, e.Identifier(), req.Model, auth)
defer reporter.trackFailure(ctx, &err)
upstreamModel := util.ResolveOriginalModel(req.Model, req.Metadata)
from := opts.SourceFormat
to := sdktranslator.FromString("codex")
body := sdktranslator.TranslateRequest(from, to, req.Model, bytes.Clone(req.Payload), false)
body = e.setReasoningEffortByAlias(req.Model, body)
body = applyReasoningEffortMetadata(body, req.Metadata, req.Model)
body = applyPayloadConfig(e.cfg, req.Model, body)
body, _ = sjson.SetBytes(body, "model", upstreamModel)
body, _ = sjson.SetBytes(body, "stream", true)
body, _ = sjson.DeleteBytes(body, "previous_response_id")
@@ -142,13 +142,16 @@ func (e *CodexExecutor) ExecuteStream(ctx context.Context, auth *cliproxyauth.Au
reporter := newUsageReporter(ctx, e.Identifier(), req.Model, auth)
defer reporter.trackFailure(ctx, &err)
upstreamModel := util.ResolveOriginalModel(req.Model, req.Metadata)
from := opts.SourceFormat
to := sdktranslator.FromString("codex")
body := sdktranslator.TranslateRequest(from, to, req.Model, bytes.Clone(req.Payload), true)
body = e.setReasoningEffortByAlias(req.Model, body)
body = applyReasoningEffortMetadata(body, req.Metadata, req.Model)
body = applyPayloadConfig(e.cfg, req.Model, body)
body, _ = sjson.DeleteBytes(body, "previous_response_id")
body, _ = sjson.SetBytes(body, "model", upstreamModel)
url := strings.TrimSuffix(baseURL, "/") + "/responses"
httpReq, err := e.cacheHelper(ctx, from, url, req, body)
@@ -235,14 +238,16 @@ func (e *CodexExecutor) ExecuteStream(ctx context.Context, auth *cliproxyauth.Au
}
func (e *CodexExecutor) CountTokens(ctx context.Context, auth *cliproxyauth.Auth, req cliproxyexecutor.Request, opts cliproxyexecutor.Options) (cliproxyexecutor.Response, error) {
upstreamModel := util.ResolveOriginalModel(req.Model, req.Metadata)
from := opts.SourceFormat
to := sdktranslator.FromString("codex")
body := sdktranslator.TranslateRequest(from, to, req.Model, bytes.Clone(req.Payload), false)
modelForCounting := req.Model
body = e.setReasoningEffortByAlias(req.Model, body)
body = applyReasoningEffortMetadata(body, req.Metadata, req.Model)
body, _ = sjson.SetBytes(body, "model", upstreamModel)
body, _ = sjson.DeleteBytes(body, "previous_response_id")
body, _ = sjson.SetBytes(body, "stream", false)
@@ -261,83 +266,6 @@ func (e *CodexExecutor) CountTokens(ctx context.Context, auth *cliproxyauth.Auth
return cliproxyexecutor.Response{Payload: []byte(translated)}, nil
}
func (e *CodexExecutor) setReasoningEffortByAlias(modelName string, payload []byte) []byte {
if util.InArray([]string{"gpt-5", "gpt-5-minimal", "gpt-5-low", "gpt-5-medium", "gpt-5-high"}, modelName) {
payload, _ = sjson.SetBytes(payload, "model", "gpt-5")
switch modelName {
case "gpt-5-minimal":
payload, _ = sjson.SetBytes(payload, "reasoning.effort", "minimal")
case "gpt-5-low":
payload, _ = sjson.SetBytes(payload, "reasoning.effort", "low")
case "gpt-5-medium":
payload, _ = sjson.SetBytes(payload, "reasoning.effort", "medium")
case "gpt-5-high":
payload, _ = sjson.SetBytes(payload, "reasoning.effort", "high")
}
} else if util.InArray([]string{"gpt-5-codex", "gpt-5-codex-low", "gpt-5-codex-medium", "gpt-5-codex-high"}, modelName) {
payload, _ = sjson.SetBytes(payload, "model", "gpt-5-codex")
switch modelName {
case "gpt-5-codex-low":
payload, _ = sjson.SetBytes(payload, "reasoning.effort", "low")
case "gpt-5-codex-medium":
payload, _ = sjson.SetBytes(payload, "reasoning.effort", "medium")
case "gpt-5-codex-high":
payload, _ = sjson.SetBytes(payload, "reasoning.effort", "high")
}
} else if util.InArray([]string{"gpt-5-codex-mini", "gpt-5-codex-mini-medium", "gpt-5-codex-mini-high"}, modelName) {
payload, _ = sjson.SetBytes(payload, "model", "gpt-5-codex-mini")
switch modelName {
case "gpt-5-codex-mini-medium":
payload, _ = sjson.SetBytes(payload, "reasoning.effort", "medium")
case "gpt-5-codex-mini-high":
payload, _ = sjson.SetBytes(payload, "reasoning.effort", "high")
}
} else if util.InArray([]string{"gpt-5.1", "gpt-5.1-none", "gpt-5.1-low", "gpt-5.1-medium", "gpt-5.1-high"}, modelName) {
payload, _ = sjson.SetBytes(payload, "model", "gpt-5.1")
switch modelName {
case "gpt-5.1-none":
payload, _ = sjson.SetBytes(payload, "reasoning.effort", "none")
case "gpt-5.1-low":
payload, _ = sjson.SetBytes(payload, "reasoning.effort", "low")
case "gpt-5.1-medium":
payload, _ = sjson.SetBytes(payload, "reasoning.effort", "medium")
case "gpt-5.1-high":
payload, _ = sjson.SetBytes(payload, "reasoning.effort", "high")
}
} else if util.InArray([]string{"gpt-5.1-codex", "gpt-5.1-codex-low", "gpt-5.1-codex-medium", "gpt-5.1-codex-high"}, modelName) {
payload, _ = sjson.SetBytes(payload, "model", "gpt-5.1-codex")
switch modelName {
case "gpt-5.1-codex-low":
payload, _ = sjson.SetBytes(payload, "reasoning.effort", "low")
case "gpt-5.1-codex-medium":
payload, _ = sjson.SetBytes(payload, "reasoning.effort", "medium")
case "gpt-5.1-codex-high":
payload, _ = sjson.SetBytes(payload, "reasoning.effort", "high")
}
} else if util.InArray([]string{"gpt-5.1-codex-mini", "gpt-5.1-codex-mini-medium", "gpt-5.1-codex-mini-high"}, modelName) {
payload, _ = sjson.SetBytes(payload, "model", "gpt-5.1-codex-mini")
switch modelName {
case "gpt-5.1-codex-mini-medium":
payload, _ = sjson.SetBytes(payload, "reasoning.effort", "medium")
case "gpt-5.1-codex-mini-high":
payload, _ = sjson.SetBytes(payload, "reasoning.effort", "high")
}
} else if util.InArray([]string{"gpt-5.1-codex-max", "gpt-5.1-codex-max-low", "gpt-5.1-codex-max-medium", "gpt-5.1-codex-max-high", "gpt-5.1-codex-max-xhigh"}, modelName) {
payload, _ = sjson.SetBytes(payload, "model", "gpt-5.1-codex-max")
switch modelName {
case "gpt-5.1-codex-max-low":
payload, _ = sjson.SetBytes(payload, "reasoning.effort", "low")
case "gpt-5.1-codex-max-medium":
payload, _ = sjson.SetBytes(payload, "reasoning.effort", "medium")
case "gpt-5.1-codex-max-high":
payload, _ = sjson.SetBytes(payload, "reasoning.effort", "high")
case "gpt-5.1-codex-max-xhigh":
payload, _ = sjson.SetBytes(payload, "reasoning.effort", "xhigh")
}
}
return payload
}
func tokenizerForCodexModel(model string) (tokenizer.Codec, error) {
sanitized := strings.ToLower(strings.TrimSpace(model))
switch {