# gpt-5.5

> LLM model from openai, available via Core.Today API.

- **Provider**: openai
- **Model ID**: gpt-5.5
- **Category**: chat
- **Pricing Type**: token_based

## API Endpoint

Base URL: https://api.core.today

### Chat Completions
POST /llm/openai/v1/chat/completions

## Authentication

Header: `Authorization: Bearer YOUR_API_KEY`

Note: Your Core.Today API key (`cdt_xxx`) works as Bearer token.

## Input Parameters

- `model` (string, **required**): Model identifier. Use `gpt-5.5`
- `messages` (array, **required**): Array of message objects with `role` and `content`
- `temperature` (number, optional): Sampling temperature (Default: `1.0`; Range: min: 0, max: 2)
- `max_tokens` (integer, optional): Maximum tokens to generate
- `stream` (boolean, optional): Enable streaming responses (Default: `false`)
- `top_p` (number, optional): Nucleus sampling parameter (Default: `1.0`)

## Example Request

```json
{
  "model": "gpt-5.5",
  "messages": [
    {
      "role": "user",
      "content": "Hello, how are you?"
    }
  ]
}
```

## Response Format

```json
{
  "id": "chatcmpl-abc123",
  "object": "chat.completion",
  "model": "gpt-5.5",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Response text here"
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 10,
    "completion_tokens": 20,
    "total_tokens": 30
  }
}
```

## Usage Flow

1. POST /llm/openai/v1/chat/completions with model and messages
2. Response is returned synchronously (or streamed if stream=true)
3. Credits are deducted based on token usage

## Token Pricing

- Input: 0.01 credits/token
- Cached Input: 0.001 credits/token
- Image Input: 0.01 credits/token (charged at the Input rate)
- Output: 0.06 credits/token

## Image Input

This model accepts images as input. Image bytes are converted to tokens by the upstream provider and charged at the standard Input rate (0.01 credits/token).

**Tokenization**: patch-based
- Tokens per image ≈ `patches × 1` (each patch = 32×32px, capped at 2500 patches)
- Notes: 32x32 patches, billed at input rate × multiplier; cap 2500 (high) / 10000 (original detail)

### Example (OpenAI-compatible)

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

## Tags

openai, chat