Chat Completions

POST /v1/chat/completions

Use the model name shown on the Pricing page.

Authentication

Authorization: Bearer OMROUTER_API_KEY
Content-Type: application/json

Request body

Field	Type	Required	Description
model	string	Yes	Model name shown on the Pricing page.
messages	array	Yes	Conversation messages in order.
messages[].role	string	Yes	Common values are system, user, assistant, and tool.
messages[].content	string or array	Yes	Text or model-supported content parts.
stream	boolean	No	Return content as it is generated when true.
max_tokens	integer	No	Maximum generated tokens where supported.
max_completion_tokens	integer	No	Newer completion-token limit where supported.
temperature	number	No	Sampling control; support and range depend on the model.
tools	array	No	Tool definitions for models that support tool calling.
response_format	object	No	Output-format constraint where supported.

Optional fields vary by model. Start with the simple request below, then add other fields one at a time.

Example

curl --silent --show-error \
  'https://omrouter.com/v1/chat/completions' \
  --header "Authorization: Bearer $OMROUTER_API_KEY" \
  --header 'Content-Type: application/json' \
  --data "{
    \"model\": \"$OMROUTER_MODEL_ID\",
    \"messages\": [
      {\"role\": \"system\", \"content\": \"Be concise.\"},
      {\"role\": \"user\", \"content\": \"What is an API gateway?\"}
    ]
  }"

Response

{
  "id": "chatcmpl-...",
  "object": "chat.completion",
  "created": 0,
  "model": "model-id",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "..."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 0,
    "completion_tokens": 0,
    "total_tokens": 0
  }
}

Some models return additional fields or slightly different token details.

Streaming

Set stream: true to receive incremental events. See Streaming responses for completion and retry rules.

Errors

If the request fails, check the API key, model name, and request fields. See Errors and retries.