400 Bad Request when reasoning_effort="xhigh" with kimi k2.5 (OpenAI-compatible API)

[kwanLeeFrmVi]

Is it a request payload issue?
[x] Yes, this is a request payload issue. I am using a client/cURL to send a request payload, but I received an unexpected error.
[ ] No, it's another issue.

Describe the bug
API Error: 400 Bad Request. Validation error for reasoning_effort: expected one of 'low' | 'medium' | 'high', received 'xhigh'.
Error snippet:

{"detail":"Invalid request: Invalid request: {\"detail\":{\"error\":{\"message\":\"1 validation error:\\n  {'type':           
  'literal_error', 'loc':                                                                                                                                           
  ('body', 'reasoning_effort'), 'msg': \\\"Input should be 'low', 'medium' or 'high'\\\", 'input': 'xhigh', 'ctx': {'expected': \\\"'low', 'medium' or              
                                                                                                                                                                    
  'high'\\\"}}\\n\\n  File \\\"/app/vllm_src/vllm/entrypoints/utils.py\\\", line 34, in create_chat_completion\\n    POST /v1/chat/completions [{'type':            
                                                                                                                                                                    
  'literal_error', 'loc': ('body', 'reasoning_effort'), 'msg': \\\"Input should be 'low', 'medium' or 'high'\\\", 'input': 'xhigh', 'ctx': {'expected':             
  \\\"'low',                                                                                                                                                        
  'medium' or 'high'\\\"}}]\",\"type\":\"Bad Request\",\"param\":null,\"code\":400}}}"} 

CLI Type
claude code

Model Name
kimi k2.5

LLM Client
claude code

OS Type

• OS: macOS (Darwin)
• Version: Darwin 25.2.0

Additional context
CLIProxyAPI v6.7.28; provider chutes (OpenAI-compatible API)

[luispater]

Kimi's reasoning effort only supports low, medium, and high.

Please set the reasoning effort to high.

[crazyn2]

Kimi's reasoning effort only supports low, medium, and high.

Please set the reasoning effort to high.

我也遇到这个情况，而且是普通的调用，不是就会出现问题，如果是anthropic 调用就会出现这个问题，如果是openai协议调用报错设置了思考模式，但是没有设置reason_effort值，也报错