Chat Completions

The chat completions endpoint creates AI responses for conversational messages. This is the primary endpoint for most applications.

Endpoint

POST https://api.elyxir.ai/v1/chat/completions

Request

Headers

Body Parameters

Not every model supports every parameter — tool calling and structured output in particular vary by provider. Check the model in the Model Hub if a parameter appears to be ignored.

Message Object

Vision

Pass an array of content parts to send images to a vision-capable model:

{
  "model": "gpt-5.5",
  "messages": [
    {
      "role": "user",
      "content": [
        {"type": "text", "text": "What is in this image?"},
        {"type": "image_url", "image_url": {"url": "https://example.com/photo.jpg"}}
      ]
    }
  ]
}

A base64 data URI works in place of an HTTPS URL.

Example Request

curl -X POST https://api.elyxir.ai/v1/chat/completions \
  -H "Authorization: Bearer elyxir_your_api_key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-4o-mini",
    "messages": [
      {"role": "system", "content": "You are a helpful assistant."},
      {"role": "user", "content": "What is the capital of France?"}
    ],
    "temperature": 0.7,
    "max_tokens": 150
  }'

Response

Success (200 OK)

{
  "id": "chatcmpl-abc123",
  "object": "chat.completion",
  "created": 1234567890,
  "model": "gpt-4o-mini",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "The capital of France is Paris."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 25,
    "completion_tokens": 10,
    "total_tokens": 35
  }
}

Response Fields

Finish Reasons

Multi-turn Conversations

Include previous messages for context:

messages = [
    {"role": "system", "content": "You are a helpful assistant."},
    {"role": "user", "content": "What is Python?"},
    {"role": "assistant", "content": "Python is a programming language..."},
    {"role": "user", "content": "How do I install it?"}
]
 
response = requests.post(
    "https://api.elyxir.ai/v1/chat/completions",
    headers=headers,
    json={"model": "gpt-4o-mini", "messages": messages}
)

Streaming

Enable streaming for real-time responses:

curl -X POST https://api.elyxir.ai/v1/chat/completions \
  -H "Authorization: Bearer elyxir_your_api_key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-4o-mini",
    "messages": [{"role": "user", "content": "Tell me a story"}],
    "stream": true
  }'

Streaming returns Server-Sent Events (SSE):

data: {"id":"chatcmpl-abc","choices":[{"delta":{"content":"The"}}]}
data: {"id":"chatcmpl-abc","choices":[{"delta":{"content":" capital"}}]}
data: [DONE]

Add "stream_options": {"include_usage": true} to receive token counts on the final chunk — otherwise a streamed response carries no usage block.

Alvin auto-routing

Set model to alvin and Alvin picks the model per request. The response reports what it chose:

The same information appears in a routing field in the response body. See Alvin routing.

Available Models

This is a shortlist. Call GET /v1/models for the live catalog, or see Supported models for the full list with pricing.

Error Responses

See Error Handling for details.

Related

• API Overview - Getting started
• Authentication - API key setup
• Memory - Persistent user and project memory