Compatibility is per model
Chat Completions is the planned compatibility surface. Check messages, tools, structured output, vision, context limits, and usage support for the exact model. Do not infer complete OpenAI compatibility from the endpoint name.
Real LLM providers are not connected in this build stage. The example documents the intended request; it must not return invented model text as a live completion.
Streaming changes recovery
- Before visible content is sent, an eligible confirmed failure may use an authorized fallback.
- After text, a tool call, or other visible SSE data is sent, do not restart on another supplier and concatenate a second answer.
- A client disconnect is not automatically free. Record known usage and resolve unknown metering according to the model’s published contract.
- Tools are instructions for your application. The platform must not repeat a tool action while recovering a stream.
Keep platform extensions explicit
If an SDK rejects custom fields such as fallback, use its supported extra-body mechanism or plain HTTP. Use only valid SDK options. The 1–4 image/video output_count field is not an LLM generation multiplier.