> ## 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.

# Edge Cases

> Error handling, timeouts, rate limits, and special scenarios.

## Authentication Errors

```json theme={"dark"}
// Missing API key (401)
{"detail": "API key is required."}

// Invalid API key (401)
{"detail": "Invalid API key. Please check your API key and try again."}
```

## Validation Errors

```json theme={"dark"}
// Missing messages (422)
{"detail": [{"loc": ["body","messages"], "msg": "field required"}]}

// Invalid image URL
{"type": "error", "message": "Invalid image URL: Unable to fetch image from provided URL"}

// Too many images
{"type": "error", "message": "Maximum 4 images allowed per request"}
```

## Model Errors

```json theme={"dark"}
// Model not found
{"type": "error", "message": "Model not found: invalid-model-name", "suggestion": "Did you mean 'nano-banana-pro'?"}

// Model unavailable
{"type": "error", "message": "Model temporarily unavailable: kling-3-0", "alternatives": ["kling-2-1"]}

// Generation failed
{"type": "error", "message": "Failed to generate image: Model returned error", "details": {"error_code": "CONTENT_FILTER"}}
```

## Timeouts

| Operation            | Timeout         |
| -------------------- | --------------- |
| HTTP connection      | 30 seconds      |
| HTTP request         | 300 seconds     |
| Streaming connection | 15 minutes idle |
| Image generation     | 10–60 seconds   |
| Video generation     | 60–600 seconds  |
| Workflow execution   | 15 minutes      |

```json theme={"dark"}
{
  "type": "error",
  "message": "Model execution timed out after 300 seconds",
  "error_code": "TIMEOUT",
  "suggestion": "For long-running operations, check execution status with the execution_id"
}
```

## Rate Limiting

```json theme={"dark"}
// Rate limit error (429)
{
  "detail": "Rate limit exceeded.",
  "error_code": "RATE_LIMITED",
  "retry_after": 30
}
```

### Limits by Plan

| Plan       | Requests/min | Concurrent |
| ---------- | ------------ | ---------- |
| Free       | 10           | 2          |
| Pro        | 60           | 10         |
| Enterprise | 300          | 50         |

### Rate Limit Headers

```
X-RateLimit-Limit: 60
X-RateLimit-Remaining: 45
X-RateLimit-Reset: 1708345678
```

## Insufficient Balance

```json theme={"dark"}
{
  "type": "error",
  "message": "Insufficient balance.",
  "error_code": "INSUFFICIENT_BALANCE",
  "details": {
    "current_balance": 0.10,
    "required_estimate": 0.50,
    "top_up_url": "https://eachlabs.ai/billing"
  }
}
```

## NSFW Content Filtering

By default, NSFW content is filtered. To disable:

```json theme={"dark"}
{
  "messages": [{"role": "user", "content": "..."}],
  "enable_safety_checker": false
}
```

<Warning>
  Only certain models support NSFW generation (e.g., Wan, Seedream). Other models ignore `enable_safety_checker: false`.
</Warning>

## Input Validation Limits

| Input         | Constraint            |
| ------------- | --------------------- |
| Images        | Max 4 per request     |
| Image formats | JPEG, PNG, WebP, GIF  |
| Image size    | Max 20MB, 64–8192px   |
| Video formats | MP4, WebM, MOV        |
| Video size    | Max 500MB, 10 min     |
| Audio formats | MP3, WAV, M4A, OGG    |
| Audio size    | Max 50MB, 5 min       |
| Prompt length | Max 10,000 characters |

## Session Edge Cases

```json theme={"dark"}
// Session busy (concurrent request)
{"type": "error", "message": "Session is currently processing another request", "error_code": "SESSION_BUSY"}

// Session memory limit
{"type": "error", "message": "Session memory limit reached. Maximum 50 messages per session."}
```

## Streaming Error Handling

Errors during streaming are emitted as events before `[DONE]`:

```
data: {"type":"status","message":"Generating image..."}
data: {"type":"error","message":"Generation failed: Insufficient balance"}
data: [DONE]
```

The stream always terminates after an error event.