diagnostic {"subsystem":"diagnostic"} lane task error: lane=session:agent:main:o...
错误信息
21:12:43+08:00 error diagnostic {"subsystem":"diagnostic"} lane task error: lane=session:agent:main:openclaw-weixin:direct:o9cq801x9sh4ed3sqfvcfs9ko-jg@im.wechat durationMs=4971 error="FailoverError: LLM request failed: provider rejected the request schema or tool payload."
21:12:43+08:00 warn model-fallback/decision {"subsystem":"model-fallback/decision"} {"event":"model_fallback_decision","tags":["error_handling","model_fallback","candidate_failed"],"runId":"fe7944a5-42b5-46d0-a71a-f4402a4c279e","sessionId":"216ec1de-2854-4c6d-847f-78bb734840d5","lane":"main","decision":"candidate_failed","requestedProvider":"custom-www-dmxapi-cn","requestedModel":"gpt-5.6-luna","candidateProvider":"custom-www-dmxapi-cn","candidateModel":"gpt-5.6-luna","attempt":1,"total":1,"reason":"format","status":400,"errorPreview":"400 Function tools with reasoning_effort are not supported for gpt-5.6-luna in /v1/chat/completions. To use function tools, use /v1/responses or set reasoning_effort to 'none'.","errorHash":"sha256:addffede8d36","fallbackStepType":"fallback_step","fallbackStepFromModel":"custom-www-dmxapi-cn/gpt-5.6-luna","fallbackStepFromFailureReason":"format","fallbackStepFromFailureDetail":"400 Function tools with reasoning_effort are not supported for gpt-5.6-luna in /v1/chat/completions. To use function tools, use /v1/responses or set reasoning_effort to 'none'.","fallbackStepChainPosition":1,"fallbackStepFinalOutcome":"chain_exhausted","isPrimary":true,"requestedModelMatched":true,"fallbackConfigured":false} model fallback decision
21:12:43+08:00 error Embedded agent failed before reply: LLM request failed: provider rejected the request schema or tool payload.
21:12:45+08:00 info gateway/channels/openclaw-weixin outbound: to=o9cq801X9sh4Ed3sqFvCFS9Ko-Jg@im.wechat contextToken=AARzJW…(len=136) textLen=110 mediaUrl=present
21:12:45+08:00 info gateway/channels/openclaw-weixin [weixin] sendWeixinMediaFile: uploading file attachment filePath=C:\Users\Administrator\.openclaw\media\outbound\voice-1790428365387---5412cefb-7659-451d-836c-5278f03a1f72.mp3 name=voice-1790428365387---5412cefb-7659-451d-836c-5278f03a1f72.mp3 to=o9cq801X9sh4Ed3sqFvCFS9Ko-Jg@im.wechat
21:12:45+08:00 info gateway/channels/openclaw-weixin [weixin] sendWeixinMediaFile: file upload done filekey=8ff9f487de445388a273457a299c78a7 size=49248
21:12:46+08:00 info gateway/channels/openclaw-weixin sendFileMessageWeixin: success to=o9cq801X9sh4Ed3sqFvCFS9Ko-Jg@im.wechat clientId=openclaw-weixin:1790428366173-9b00de56
21:12:46+08:00 info gateway/channels/openclaw-weixin outbound: media sent OK to=o9cq801X9sh4Ed3sqFvCFS9Ko-Jg@im.wechat
21:12:46+08:00 info gateway/channels/openclaw-weixin debug-check: accountId=6b561eb4c6d7-im-bot debug=false hasContextToken=true
AI 解决方案
🔧 报错分析报告
1. 🌐 错误翻译
原文：400 Function tools with reasoning_effort are not supported for gpt-5.6-luna in /v1/chat/completions. To use function tools, use /v1/responses or set reasoning_effort to 'none'.
译文：错误码400：在 /v1/chat/completions 接口中，GPT-5.6-luna 模型不支持同时使用 Function Tools（函数工具）和 reasoning_effort 参数。若要使用函数工具，请改用 /v1/responses 接口，或将 reasoning_effort 参数设置为 'none'。
2. 🎯 原因分析
/v1/chat/completions 接口调用模型 gpt-5.6-luna 时，请求payload同时包含了“Function Tools（函数工具）”和“reasoning_effort（推理强度）”参数，该模型不支持这种配置组合，直接拒绝并返回 400 错误。3. 💡 解决方案
请调整发起请求的代码，按以下两种方案之一修改：
方案A：移除 reasoning_effort 参数（推荐）
// ❌ 错误代码
const response = await openai.chat.completions.create({
model: "gpt-5.6-luna",
reasoning_effort: "medium", // 冲突参数
tools: [myFunctionTool]
});
// ✅ 方案A：移除冲突参数
const response = await openai.chat.completions.create({
model: "gpt-5.6-luna",
// 删除 reasoning_effort 参数
tools: [myFunctionTool]
});
方案B：改用 /v1/responses 接口
// ✅ 方案B：切换到兼容接口，同时保留参数
const response = await openai.responses.create({
model: "gpt-5.6-luna",
reasoning_effort: "medium", // 在新接口中兼容
tools: [myFunctionTool]
});