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

# Model Aliases

> Use simple aliases like 'best', 'fast', 'code' instead of memorizing model IDs.

Don't know which model to pick? Use an alias. Kyma resolves it to the best model for that category — and auto-updates when better models become available.

## Available Aliases

| Alias          | Resolves To        | Best For                                                       |
| -------------- | ------------------ | -------------------------------------------------------------- |
| `best`         | `qwen-3.6-plus`    | Highest quality overall                                        |
| `fast`         | `qwen-3-32b`       | Fastest active coding / response path                          |
| `code`         | `qwen-3-coder`     | Code-focused tasks                                             |
| `cheap`        | `gemini-2.5-flash` | Lowest-cost active alias                                       |
| `long-context` | `gemini-2.5-flash` | 1M token context window                                        |
| `vision`       | `gemma-4-31b`      | Image input / multimodal                                       |
| `reasoning`    | `deepseek-r1`      | Complex reasoning tasks                                        |
| `agent`        | `kimi-k2.6`        | Agentic tool use workflows                                     |
| `best-agent`   | `kimi-k2.6`        | Agentic tool use workflows (explicit alias)                    |
| `balanced`     | `llama-3.3-70b`    | Good quality + speed balance                                   |
| `glm-flagship` | `glm-5.2`          | Long-running coding agents, repo-scale engineering, 1M context |
| `search`       | `sonar`            | Live web search with citations (current events, research)      |

## Audio Aliases

Kyma also exposes shortcuts for the audio stack. Use them with `/v1/audio/transcriptions` or `/v1/audio/understand` instead of memorising provider model IDs.

| Alias                | Resolves To                         | Best For                                                                               |
| -------------------- | ----------------------------------- | -------------------------------------------------------------------------------------- |
| `transcribe`         | `whisper-v3-turbo`                  | Speech-to-text, captions, voice agents (228× realtime)                                 |
| `transcribe-quality` | `gpt-4o-mini-transcribe-2025-12-15` | High-accuracy transcription, multilingual code-switching (Vi/En), conversational audio |
| `audio-understand`   | `gemini-3-flash-audio`              | Scene/tone/music recognition beyond transcription                                      |

```bash theme={null}
# Transcribe with the alias
curl https://kymaapi.com/v1/audio/transcriptions \
  -H "Authorization: Bearer $KYMA_KEY" \
  -F "file=@meeting.mp3" \
  -F "model=transcribe"

# Audio scene understanding
curl https://kymaapi.com/v1/audio/understand \
  -H "Authorization: Bearer $KYMA_KEY" \
  -F "file=@clip.mp3" \
  -F "question=What instruments and mood?" \
  -F "model=audio-understand"
```

For TTS (`/v1/audio/speech`), music (`/v1/audio/music`), SFX (`/v1/audio/sfx`), and voice ops (`/v1/audio/voice-clone`, `/v1/audio/voice-design`), pass the explicit model id (e.g. `eleven-multilingual-v2`, `minimax-music-pro`). Aliases for those will land once we settle on canonical defaults.

## Usage

Use aliases anywhere you'd use a model ID:

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

client = OpenAI(base_url="https://kymaapi.com/v1", api_key="YOUR_KEY")

# Instead of remembering "qwen-3.6-plus"
response = client.chat.completions.create(
    model="best",  # → resolves to qwen-3.6-plus
    messages=[{"role": "user", "content": "Hello!"}]
)
```

```bash theme={null}
# cURL
curl https://kymaapi.com/v1/chat/completions \
  -H "Authorization: Bearer YOUR_KEY" \
  -d '{"model": "code", "messages": [{"role": "user", "content": "Fix this bug"}]}'
```

## In Coding Agents

<Tabs>
  <Tab title="Cline">
    Set Model ID to `code` in Cline settings.
  </Tab>

  <Tab title="OpenClaw">
    ```json theme={null}
    { "primary": "kyma/code" }
    ```
  </Tab>

  <Tab title="Aider">
    ```bash theme={null}
    aider --model openai/best --openai-api-base https://kymaapi.com/v1
    ```
  </Tab>

  <Tab title="Cursor">
    Set model to `best` in Cursor's OpenAI-compatible provider settings.
  </Tab>
</Tabs>

## Transparency

When you use an alias, Kyma adds headers so you always know which model was used:

```
X-Kyma-Model: qwen-3.6-plus    # actual model used
X-Kyma-Alias: best              # alias you sent
```

## Auto-Updates

Aliases are updated when better models become available. For example, if a new model outperforms `qwen-3.6-plus`, the `best` alias will point to it automatically — no config changes needed.

If you need a **fixed** model that never changes, use the full model ID (e.g., `qwen-3.6-plus`) instead of an alias.

## Get Current Aliases

```bash theme={null}
curl https://kymaapi.com/v1/models | jq '.aliases'
```

```json theme={null}
{
  "best": "qwen-3.6-plus",
  "fast": "qwen-3-32b",
  "code": "qwen-3-coder",
  "cheap": "gemini-2.5-flash",
  "long-context": "gemini-2.5-flash",
  "vision": "gemma-4-31b",
  "reasoning": "deepseek-r1",
  "agent": "kimi-k2.6",
  "best-agent": "kimi-k2.6",
  "balanced": "llama-3.3-70b",
  "glm-flagship": "glm-5.2",
  "search": "sonar",
  "transcribe": "whisper-v3-turbo",
  "transcribe-quality": "gpt-4o-mini-transcribe-2025-12-15",
  "audio-understand": "gemini-3-flash-audio"
}
```

## Smart Recommendations

Not sure which alias fits? Use the recommend API:

```bash theme={null}
# "I use Cline, what model should I use?"
curl "https://kymaapi.com/v1/models/recommend?agent=cline"

# "I need a coding model with tool support"
curl "https://kymaapi.com/v1/models/recommend?for=coding"
```

Supported agents: `cline`, `roo-code`, `openclaw`, `claude-code`, `aider`, `opencode`, `cursor`, `kilo-code`.

See [Models](/guides/models) for full model details.