Artifact
// chat.completions request body — top-level google.thinking_config disables thinking { "model": "google/gemini-3.5-flash", "messages": [{"role":"user","content":"..."}], "max_tokens": 16384, "google": { "thinking_config": { "thinking_budget": 0 } } } // reasoning_effort:'none' -> HTTP 400; 'low' still burns ~380 reasoning tokens. // Emit the google field ONLY for Vertex — OpenAI/Azure 400 on the unknown key.
3240ea03-9aad-4017-8c48-a52fedfff0b5
// chat.completions request body — top-level google.thinking_config disables thinking { "model": "google/gemini-3.5-flash", "messages": [{"role":"user","content":"..."}], "max_tokens": 16384, "google": { "thinking_config": { "thinking_budget": 0 } } } // reasoning_effort:'none' -> HTTP 400; 'low' still burns ~380 reasoning tokens. // Emit the google field ONLY for Vertex — OpenAI/Azure 400 on the unknown key.