Create a chat completion (streaming or not)

GuardrailsPrompt templatesA/B testingCredit meteringCaching

Authentication

Send your virtual key as Authorization: Bearer sk-nemo-…. One key works across every model and modality.

Request bodyapplication/json

Field	Type	Required
model	string	required
messages	array	required
temperature	number	optional
max_tokens	integer	optional
n	integer	optional
stream	boolean	optional
response_format	object	optional
tools	array	optional
tool_choice	object	optional

Response200

{
  "id": "chatcmpl-abc123",
  "object": "chat.completion",
  "model": "gpt-4o",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Hello!"
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 9,
    "completion_tokens": 2,
    "total_tokens": 11
  }
}

Errors

Status	Meaning
400	400 — Guardrail blocked the request
402	402 — Not enough credits
429	429 — one of: rate_limit_exceeded (RPM limit exceeded); tpm_limit_exceeded (TPM limit exceeded)
503	503 — Credit system unavailable