> ## Documentation Index
> Fetch the complete documentation index at: https://docs.eachlabs.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Chat Completions

> OpenAI-compatible chat completions endpoint. The primary interface for all each::sense interactions.

## Endpoint

```
POST https://eachsense-agent.core.eachlabs.run/v1/chat/completions
```

## Request Body

```typescript theme={"dark"}
interface ChatCompletionRequest {
  // Required
  messages: Array<{
    role: "user" | "assistant" | "system";
    content: string;
  }>;

  // OpenAI compatible
  model?: string;              // Default: "eachsense/beta"
  stream?: boolean;            // Default: true

  // each::labs extensions
  session_id?: string;         // Conversation continuity
  mode?: "max" | "eco";        // Quality mode (default: "max")
  behavior?: "agent" | "plan" | "ask";  // Behavior mode
  image_urls?: string[];       // Image inputs (max 4)
  workflow_id?: string;        // For workflow operations
  version_id?: string;         // Workflow version
  web_search?: boolean;        // Enable web search (default: true)
  enable_safety_checker?: boolean;  // NSFW filter (default: true)
}
```

### Parameters

| Parameter               | Type      | Required | Default            | Description                             |
| ----------------------- | --------- | -------- | ------------------ | --------------------------------------- |
| `messages`              | array     | Yes      | `-`                | OpenAI-compatible message array         |
| `model`                 | string    | No       | `"eachsense/beta"` | Model identifier                        |
| `stream`                | boolean   | No       | `true`             | Enable SSE streaming                    |
| `session_id`            | string    | No       | `-`                | Session ID for conversation continuity  |
| `mode`                  | string    | No       | `"max"`            | Quality mode: `"max"` or `"eco"`        |
| `behavior`              | string    | No       | `"agent"`          | `"agent"`, `"plan"`, or `"ask"`         |
| `image_urls`            | string\[] | No       | `-`                | Image URLs for analysis/editing (max 4) |
| `workflow_id`           | string    | No       | `-`                | Target workflow for execution           |
| `version_id`            | string    | No       | `-`                | Workflow version                        |
| `web_search`            | boolean   | No       | `true`             | Enable web search capability            |
| `enable_safety_checker` | boolean   | No       | `true`             | NSFW content filter                     |

## Code Examples

<CodeGroup>
  ```bash cURL theme={"dark"}
  curl -X POST https://eachsense-agent.core.eachlabs.run/v1/chat/completions \
    -H "Content-Type: application/json" \
    -H "X-API-Key: YOUR_API_KEY" \
    -d '{
      "messages": [{"role": "user", "content": "Generate a sunset over mountains"}],
      "stream": true,
      "session_id": "my-session",
      "mode": "max"
    }'
  ```

  ```python Python theme={"dark"}
  from openai import OpenAI

  client = OpenAI(
      api_key="YOUR_API_KEY",
      base_url="https://eachsense-agent.core.eachlabs.run/v1"
  )

  response = client.chat.completions.create(
      model="eachsense/beta",
      messages=[{"role": "user", "content": "Generate a sunset over mountains"}],
      stream=False
  )

  print(response.choices[0].message.content)
  ```

  ```javascript JavaScript theme={"dark"}
  import OpenAI from "openai";

  const client = new OpenAI({
    apiKey: "YOUR_API_KEY",
    baseURL: "https://eachsense-agent.core.eachlabs.run/v1",
  });

  const response = await client.chat.completions.create({
    model: "eachsense/beta",
    messages: [{ role: "user", content: "Generate a sunset over mountains" }],
    stream: false,
  });

  console.log(response.choices[0].message.content);
  ```
</CodeGroup>

## Response (Non-Streaming)

```json theme={"dark"}
{
  "id": "chatcmpl-abc123",
  "object": "chat.completion",
  "created": 1708345678,
  "model": "eachsense/beta",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Here's your generated image of a sunset over mountains!"
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 15,
    "completion_tokens": 25,
    "total_tokens": 40
  },
  "generations": ["https://storage.eachlabs.ai/xxx.png"],
  "task_id": "task_abc123",
  "session_id": "my-session"
}
```

### Extended Response Fields

| Field         | Type      | Description               |
| ------------- | --------- | ------------------------- |
| `generations` | string\[] | URLs of generated media   |
| `task_id`     | string    | Unique task identifier    |
| `session_id`  | string    | Session ID for continuity |
| `workflow_id` | string    | Created/used workflow ID  |

## Response (Streaming)

See [Streaming](/sense/streaming/overview) for the full SSE streaming format with 18 event types.

```
data: {"id":"chatcmpl-xxx","object":"chat.completion.chunk","model":"eachsense/beta","choices":[{"index":0,"delta":{"content":"I'll generate..."},"finish_reason":null}],"eachlabs":{"type":"text_response"}}

data: {"id":"chatcmpl-xxx","object":"chat.completion.chunk","model":"eachsense/beta","choices":[{"index":0,"delta":{},"finish_reason":"stop"}],"eachlabs":{"type":"generation_response","url":"https://..."}}

data: [DONE]
```

## Error Responses

| Status | Body                                                                  | Description      |
| ------ | --------------------------------------------------------------------- | ---------------- |
| `401`  | `{"detail": "API key is required."}`                                  | Missing API key  |
| `401`  | `{"detail": "Invalid API key."}`                                      | Invalid API key  |
| `422`  | `{"detail": [{"loc": ["body","messages"], "msg": "field required"}]}` | Missing messages |
| `429`  | `{"detail": "Rate limit exceeded."}`                                  | Rate limited     |