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

# Models Catalog

> Complete list of available AI models available through the Emby API.

# Models Catalog

Emby lets you access multiple open-source and commercial models through one simple, OpenAI-compatible API.

Every model has a **model ID** — use this ID inside your API calls.

***

## Live Models Data

> 🔄 *Interactive model list coming soon.*\
> For now, refer to the tables below.

***

# Model Selection Guide

## By Use Case

### Fast Responses

* `kimi-k1.5-mini`
* `qwen2.5-coder-mini`
* `deepseek-v3-mini`
* `glm-4-air-mini`

### Complex Reasoning

* `kimi-k2`
* `deepseek-v3`
* `glm-4.1`
* `qwen3.5-coder-large`

### Cost-Effective

* `deepseek-v3-mini`
* `qwen2.5-coder-mini`
* `glm-4-air-mini`

### Large Context Windows

* `qwen3.5-coder-large-128k`
* `glm-4.1-200k`
* `deepseek-r1-128k`

### Vision Models

* `glm-4v`
* `qwen2.5-vision`
* `deepseek-v3-vision`

### Code Generation

* `kimi-k2`
* `deepseek-v3`
* `qwen3.5-coder-large`

***

# By Budget

| Budget Tier    | Recommended Models                    | Use Cases                          |
| -------------- | ------------------------------------- | ---------------------------------- |
| **Economy**    | deepseek-v3-mini, qwen-2.5-coder-mini | Simple tasks, prototyping          |
| **Standard**   | kimi-k1.5-pro, glm-4-air              | Production coding assistants       |
| **Premium**    | kimi-k2, qwen-3.5-coder-large         | Reasoning, refactoring             |
| **Enterprise** | deepseek-v3-128k, glm-4.1-200k        | Complex systems, RAG, long context |

***

# Using Models in Code

Use the model **ID** in all Emby API requests.

***

## Python

```python theme={null}
from openai import OpenAI

client = OpenAI(
    base_url="https://api.emby.dev/v1",
    api_key="your-emby-key"
)

# Kimi K2
resp = client.chat.completions.create(
    model="kimi-k2",
    messages=[{"role": "user", "content": "Hello!"}]
)

# DeepSeek V3
resp = client.chat.completions.create(
    model="deepseek-v3",
    messages=[{"role": "user", "content": "Explain async in Python"}]
)

# Qwen 3.5 Coder
resp = client.chat.completions.create(
    model="qwen3.5-coder-large",
    messages=[{"role": "user", "content": "Refactor this code"}]
)
```

***

## JavaScript

```javascript theme={null}
const response = await fetch("https://api.emby.dev/v1/chat/completions", {
  method: "POST",
  headers: {
    "Authorization": `Bearer ${process.env.EMBY_API_KEY}`,
    "Content-Type": "application/json"
  },
  body: JSON.stringify({
    model: "kimi-k2",
    messages: [{ role: "user", content: "Hello!" }]
  })
});

console.log(await response.json());
```

***

## cURL

```bash theme={null}
curl https://api.emby.dev/v1/chat/completions \
  -H "Authorization: Bearer $EMBY_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-v3",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'
```

***

# Automatic Fallback

Use model fallback to ensure requests never fail:

```python theme={null}
client.chat.completions.create(
    model="kimi-k2",
    messages=messages,
    fallback_models=["deepseek-v3", "qwen3.5-coder-large"],
    fallback_on_rate_limit=True,
    fallback_on_error=True,
)
```

***