# Claude Sonnet 4.6 API

Complete API reference for Claude Sonnet 4.6 LLM models powered by Anthropic.

## Model Variants

This API supports two Claude Sonnet 4.6 model variants:

| Model | ID | Description |
|-------|----|-------------|
| **Claude Sonnet 4.6 Text to Text** | `claude-sonnet-4.6-text-to-text` | Fast text reasoning and writing |
| **Claude Sonnet 4.6 File Analysis** | `claude-sonnet-4.6-file-analysis` | File and image analysis with Claude Sonnet |

## Quick Comparison

| Feature | Text to Text | File Analysis |
|---------|--------------|---------------|
| Text input | Required ✅ | Optional |
| Image input | Not supported ❌ | Optional |
| File input | Not supported ❌ | Optional |
| Streaming response | SSE | SSE |
| `max_tokens` | Optional | Optional |
| `top_p` | Optional | Optional |
| `top_k` | Optional | Optional |

## Claude Sonnet 4.6 Text to Text

### Endpoint

```http
POST /api/v1/chat/completions
```

### Request Parameters

#### Required

| Parameter | Type | Description |
|-----------|------|-------------|
| `model` | string | Must be `"claude-sonnet-4.6-text-to-text"` |
| `messages` | array | Ordered messages. See LLM API for the message structure. |
| `stream` | boolean | `true` for SSE, `false` for a single JSON response. |

#### Message Support

| Field | Text | File | Image |
|-------|------|------|-------|
| `messages` | Required ✅ | Not supported ❌ | Not supported ❌ |

#### Optional

| Parameter | Type | Range | Description |
|-----------|------|-------|-------------|
| `max_tokens` | integer | `1` - `64000` | Maximum output tokens. |
| `top_p` | number | `0.0` - `1.0` | Limits cumulative probability mass for candidate tokens. |
| `top_k` | integer | `> 0` | Limits the number of candidate tokens. |

### Example Request

```javascript
const response = await fetch('https://api.flaq.ai/api/v1/chat/completions', {
  method: 'POST',
  headers: {
    'Authorization': 'Bearer YOUR_API_KEY',
    'Accept': 'text/event-stream',
    'Content-Type': 'application/json'
  },
  body: JSON.stringify({
    model: 'claude-sonnet-4.6-text-to-text',
    messages: [
      {
        role: 'user',
        content: 'Write a concise product update for a developer audience.'
      }
    ],
    stream: true,
    max_tokens: 2048,
    top_p: 0.5,
    top_k: 50
  })
});
```

## Claude Sonnet 4.6 File Analysis

### Endpoint

```http
POST /api/v1/chat/completions
```

### Request Parameters

#### Required

| Parameter | Type | Description |
|-----------|------|-------------|
| `model` | string | Must be `"claude-sonnet-4.6-file-analysis"` |
| `messages` | array | Ordered messages. See LLM API and Message Attachments. |
| `stream` | boolean | `true` for SSE, `false` for a single JSON response. |

#### Message Support

| Field | Text | File | Image |
|-------|------|------|-------|
| `messages` | Optional | Optional | Optional |

#### Optional

| Parameter | Type | Range | Description |
|-----------|------|-------|-------------|
| `max_tokens` | integer | `1` - `64000` | Maximum output tokens. |
| `top_p` | number | `0.0` - `1.0` | Limits cumulative probability mass for candidate tokens. |
| `top_k` | integer | `> 0` | Limits the number of candidate tokens. |

### Example Request

```javascript
const response = await fetch('https://api.flaq.ai/api/v1/chat/completions', {
  method: 'POST',
  headers: {
    'Authorization': 'Bearer YOUR_API_KEY',
    'Accept': 'text/event-stream',
    'Content-Type': 'application/json'
  },
  body: JSON.stringify({
    model: 'claude-sonnet-4.6-file-analysis',
    messages: [
      {
        role: 'user',
        content: [
          { type: 'text', text: 'Summarize this file and extract the key risks.' },
          {
            type: 'file',
            file: {
              filename: 'demo.pdf',
              file_data: 'https://example.com/demo.pdf'
            }
          }
        ]
      }
    ],
    stream: true,
    max_tokens: 4096,
    top_p: 0.5,
    top_k: 50
  })
});
```

## Response Format

Claude Sonnet 4.6 LLM models return OpenAI-compatible completion responses. With `stream: true`, the response is Server-Sent Events; with `stream: false`, the response is a single JSON object.

### Successful Response

```text
data: {"id":"chatcmpl-...","object":"chat.completion.chunk","created":1710000000,"model":"claude-sonnet-4.6-file-analysis","choices":[{"index":0,"delta":{"role":"assistant"},"finish_reason":null}]}

data: {"id":"chatcmpl-...","object":"chat.completion.chunk","created":1710000000,"model":"claude-sonnet-4.6-file-analysis","choices":[{"index":0,"delta":{"content":"Here is a concise summary"},"finish_reason":null}]}

data: {"id":"chatcmpl-...","object":"chat.completion.chunk","created":1710000000,"model":"claude-sonnet-4.6-file-analysis","choices":[{"index":0,"delta":{},"finish_reason":"stop"}]}

data: {"id":"chatcmpl-...","object":"chat.completion.chunk","created":1710000000,"model":"claude-sonnet-4.6-file-analysis","choices":[],"usage":{"prompt_tokens":12,"completion_tokens":8,"total_tokens":20}}

data: [DONE]
```

### Error Response

```text
event: error
data: {"error":{"message":"API requests too frequent, exceeding rate limit","type":"rate_limit_error","code":"1302","param":null}}
```

## Best Practices

1. **Use structured messages:** Send conversation history through `messages[]` instead of flattening context into one prompt.
2. **Choose the right model variant:** Use Text to Text for text-only reasoning and writing, and File Analysis when files or images are needed.
3. **Tune Claude parameters carefully:** `top_p` and `top_k` are supported and can change output diversity.
4. **Consume SSE events:** Append `choices[0].delta.content` for streaming display and treat `data: [DONE]` as successful completion.
5. **Send only supported inputs:** Text models ignore unsupported file and image parts in the current client.