> ## 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.

# Roo Code

> Use Kyma with Roo Code VS Code extension.

Roo Code is an AI dev team in your IDE. Configure Kyma as an OpenAI Compatible provider.

## Setup

1. Open Roo Code settings in VS Code
2. Select **"OpenAI Compatible"** as the provider
3. Fill in:

| Field        | Value                    |
| ------------ | ------------------------ |
| **Base URL** | `https://kymaapi.com/v1` |
| **API Key**  | Your `ky-` key           |
| **Model ID** | `llama-3.3-70b`          |

## Recommended Models

| Model               | Alias      | Best For                                                              |
| ------------------- | ---------- | --------------------------------------------------------------------- |
| `qwen-3.6-plus`     | `best`     | **Recommended** — highest quality, #1 on Kyma                         |
| `qwen-3-coder`      | `code`     | Best alias for code-focused tasks                                     |
| `kimi-k2.6`         | `agent`    | Best tool calling, 262K context                                       |
| `llama-3.3-70b`     | `balanced` | Great all-rounder                                                     |
| `deepseek-v4-pro`   | —          | 1M context, top reasoning. Complex refactors and long-codebase tasks. |
| `deepseek-v4-flash` | —          | 1M context, value tier. Everyday completions where speed matters.     |

```bash theme={null}
# Get personalized recommendation
curl "https://kymaapi.com/v1/models/recommend?agent=roo-code"
```

See all [model aliases](/guides/model-aliases) for shortcuts.

## Audio Models (optional)

Roo Code is a coding agent, but if you build voice-enabled extensions or transcript pipelines on the side, Kyma's audio aliases drop in like any chat model:

* `transcribe` → `whisper-v3-turbo` — 228× realtime speech-to-text
* `audio-understand` → `gemini-3-flash-audio` — scene/tone/music recognition

Use them via `POST /v1/audio/transcriptions` and `POST /v1/audio/understand`. The full audio surface (TTS, music, SFX, voice clone/design) is documented in the [Audio API reference](/api-reference/audio-speech) and [Model Aliases](/guides/model-aliases#audio-aliases).

## Prompt Caching

Prompt caching is automatic. Your system prompt and tool definitions are cached for 5 minutes, reducing costs by up to 90% on subsequent requests. No code changes needed. See [Prompt Caching](/guides/prompt-caching) for details.