> ## 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.

# each::labs Documentation

> The generative media platform. 500+ AI models, multi-step workflows, and an intelligent agent interface. Build with each::api, each::workflows, and each::sense.

each::labs is a generative media platform for developers and AI agents. Get access to 500+ production-ready AI models with just one API key.

## Products

<CardGroup cols={2}>
  <Card title="each::api" icon="bolt" href="/api/overview">
    Direct model access. Run image, video, audio, and 3D generation models from a single endpoint. Async predictions with webhook delivery.
  </Card>

  <Card title="each::workflows" icon="diagram-project" href="/workflows/overview">
    Multi-step pipelines. Chain models together with parameter passing, versioning, fallback configs, and bulk execution.
  </Card>

  <Card title="each::sense" icon="brain" href="/sense/overview">
    AI agent interface. OpenAI-compatible endpoint that understands natural language, auto-selects models, generates media, and builds workflows.
  </Card>

  <Card title="LLM Router" icon="route" href="/llm-router/overview">
    One endpoint, 300+ LLMs. OpenAI, Anthropic, Google, and more. OpenAI-compatible, just swap the base URL.
  </Card>

  <Card title="MCP Server" icon="plug" href="/mcp">
    Connect to Cursor, Claude Desktop, Windsurf, and more. Generate media without ever leaving your IDE.
  </Card>
</CardGroup>

## Quick Start

<Steps>
  <Step title="Get your API key">
    Sign up at [eachlabs.ai](https://eachlabs.ai) and create an API key from [Settings → API Keys](https://eachlabs.ai/settings/api-keys).
  </Step>

  <Step title="Generate an image">
    <CodeGroup>
      ```bash cURL theme={"dark"}
      curl -X POST https://api.eachlabs.ai/v1/prediction \
        -H "Content-Type: application/json" \
        -H "X-API-Key: $EACHLABS_API_KEY" \
        -d '{
          "model": "nano-banana-pro",
          "input": {
            "prompt": "A cinematic landscape at golden hour",
            "aspect_ratio": "16:9"
          }
        }'
      ```

      ```python Python theme={"dark"}
      import requests

      response = requests.post(
          "https://api.eachlabs.ai/v1/prediction",
          headers={"X-API-Key": "YOUR_API_KEY"},
          json={
              "model": "nano-banana-pro",
              "input": {
                  "prompt": "A cinematic landscape at golden hour",
                  "aspect_ratio": "16:9"
              }
          }
      )

      prediction_id = response.json()["predictionID"]
      ```

      ```javascript JavaScript theme={"dark"}
      const response = await fetch("https://api.eachlabs.ai/v1/prediction", {
        method: "POST",
        headers: {
          "Content-Type": "application/json",
          "X-API-Key": "YOUR_API_KEY",
        },
        body: JSON.stringify({
          model: "nano-banana-pro",
          input: {
            prompt: "A cinematic landscape at golden hour",
            aspect_ratio: "16:9",
          },
        }),
      });

      const { predictionID } = await response.json();
      ```
    </CodeGroup>
  </Step>

  <Step title="Or use the AI agent">
    Skip model selection entirely and let each::sense handle it for you:

    ```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 cinematic landscape at golden hour"}],
    )
    ```
  </Step>
</Steps>

## Base URLs

| Product         | Base URL                                    | Protocol   | Status                                                                    |
| --------------- | ------------------------------------------- | ---------- | ------------------------------------------------------------------------- |
| each::api       | `https://api.eachlabs.ai`                   | REST       | Canonical api-service host                                                |
| each::workflows | `https://workflows.eachlabs.run/api/v1`     | REST       | Workflows-engine host; planned to route through `https://api.eachlabs.ai` |
| each::sense     | `https://eachsense-agent.core.eachlabs.run` | REST + SSE | Current each::sense host                                                  |
| LLM Router      | `https://api.eachlabs.ai/v1`                | REST + SSE | Canonical api-service host                                                |

## For AI Agents

each::sense is fully OpenAI-compatible. Just point any agent framework at the base URL with your API key and you're good to go:

```python theme={"dark"}
# Works with LangChain, CrewAI, AutoGen, or any OpenAI-compatible client
client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://eachsense-agent.core.eachlabs.run/v1"
)
```

Machine-readable documentation index available at [`/llms.txt`](/llms.txt).

<CardGroup cols={2}>
  <Card title="Authentication" icon="key" href="/authentication">
    API key setup and security best practices.
  </Card>

  <Card title="Error Reference" icon="triangle-exclamation" href="/errors">
    HTTP status codes and error handling across all products.
  </Card>
</CardGroup>