Streaming.

GET  /v1/models                  model listing with pricing, context, features
GET  /v1/models/{id}             one model by instance ID or model ID
POST /chat                       OpenAI-compatible chat completions
POST /chat-stream                Anthropic-style SSE rich-shape
POST /v1/chat/completions        OpenAI streaming SSE

tenzro_chat                      simple shape
tenzro_chatStream                rich shape with reasoning, tool calls

Streaming responses always include usage — no stream_options.include_usage opt-in. The stream ends with a finish_reason chunk, then a usage chunk with empty choices carrying prompt_tokens / completion_tokens / total_tokens plus extension fields cost_wei, generation_time_ms, and tokens_per_second, then data: [DONE]. Streams resume via the SSE Last-Event-ID header (<completion_id>:<seq>) against an in-memory buffer with a 300-second TTL.

Event bus subscriptions for inference completion, settlement, and provider status. Replay supported via since=<height>.