chat

Chat Stream

Stream chat response via Server-Sent Events.

Events:

  • {type: "meta", chunks_used, rag_time_ms} — sent once after RAG search
  • {type: "token", text} — one per LLM token
  • {type: "done", model, provider, llm_time_ms, total_time_ms} — final
  • :keepalive — comment heartbeat every 15 s
  • [DONE] — stream terminator
post/v1/chat/stream

Request

  • Base URL: https://api.bitpoort.com (the document declares no server; this is the origin it was published from)
  • URL: https://api.bitpoort.com/v1/chat/stream
  • Auth: one of:
    • API key in header X-API-Key
    • HTTP bearer

Request body

messagestring required

The user's question

chainstring nullable

Filter context to a specific chain (e.g. 'ETH')

conversation_idstring nullable

Client-generated conversation id. On the bridge engine it maps to a live managed-agent session (true multi-turn: prior tool results stay in the agent's context). The legacy engine ignores it.

max_chunksinteger

Max RAG chunks to retrieve

providerstring nullable

Force a provider: 'anthropic' or 'deepseek'. None = auto-detect.

modelstring nullable

Per-request model override (e.g. 'claude-sonnet-4-20250514', 'gpt-4o'). If set, takes priority over server config.

llm_api_keystring nullable

Per-request LLM API key override — bring-your-own-key for hybrid RPC synthesis.

Response

Successful Response

unknown required

Changes

No recorded changes to this operation across all 1 revision of this API.