POST /api/chat — dat.ai Docs

Ollama-compatible chat. Same request/response shape as Ollama /api/chat . The response arrives in the same HTTP request.

Endpoint


POST https://llm.dat.ai/api/chat
Authorization: Bearer YOUR_API_KEY
Content-Type: application/json

Request body

Uses the Ollama chat schema . Full request — see Example below.

Field	Required	Description
`model`	Yes	Model name, e.g. `qwen3:1.7b`
`messages`	Yes	Chat history — see `messages` below
`stream`	No	Default `false` if omitted. Set `true` for streaming
`tools`	No	Ollama function calling — passed through; not the same as `datai.tools`
`datai`	No	dat.ai built-in node tools (`net`, `fs`) — see Inference overview

`messages`

Array of objects with role and content:


"messages": [
  { "role": "system", "content": "You are a helpful assistant." },
  { "role": "user", "content": "Hello!" }
]

Roles: user, assistant, system (same as Ollama /api/chat ).

Other Ollama /api/chat fields (options, etc.) are passed through as-is.

Optional header: X-Request-ID.

Response

Non-streaming (200): Ollama ChatResponse with the final message in message.content and "done": true.

Streaming: Content-Type: application/x-ndjson — one JSON object per line, each with partial message.content, ending with a final object where "done": true.

Errors

Plain JSON { "error": "..." } — see Authentication. Common cases:

Status	Cause
400	Invalid JSON, missing `model`, `datai.tools` + `stream: true`
401	Missing or invalid API key
402	Insufficient credits
504	Request timed out (max wait 180 seconds)

Example


curl -sS -X POST "https://llm.dat.ai/api/chat" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "qwen3:1.7b",
    "messages": [{"role": "user", "content": "Hello!"}],
    "stream": false
  }'