> ## 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.

# Continue.dev

> Use Kyma with Continue.dev IDE extension.

Continue is an open-source AI code assistant for VS Code and JetBrains. Configure Kyma as an OpenAI-compatible provider.

## config.yaml (recommended)

Open Continue settings → edit `config.yaml`:

```yaml theme={null}
name: My Config
version: 0.0.1
schema: v1

models:
  - name: Kyma - MiniMax M2.5
    provider: openai
    model: minimax-m2.5
    apiBase: https://kymaapi.com/v1
    apiKey: kyma-your-api-key
    roles:
      - chat
      - edit
      - apply

  - name: Kyma - Qwen 3 32B (fast)
    provider: openai
    model: qwen-3-32b
    apiBase: https://kymaapi.com/v1
    apiKey: kyma-your-api-key
    roles:
      - autocomplete
```

## config.json (legacy)

```json theme={null}
{
  "models": [
    {
      "title": "Kyma - Llama 3.3 70B",
      "provider": "openai",
      "model": "llama-3.3-70b",
      "apiBase": "https://kymaapi.com/v1",
      "apiKey": "kyma-your-api-key"
    }
  ]
}
```

## Recommended Models

| Use Case | Model | Why | Free tier |
| - | - | - | - |
| Coding | `minimax-m2.5` | Top agentic coding model on Kyma | Yes |
| Chat & Edit | `qwen-3.6-plus` | Highest overall quality | Yes |
| Autocomplete | `qwen-3-32b` | Fastest active coding model | Yes |
| Complex tasks | `deepseek-r1` | Strongest reasoning | No, needs a top-up |
| Everyday completions | `deepseek-v4-flash` | 1M context, value tier, fast | Yes |

A new account's signup credit is added once your email is verified and spends on [free tier](/pricing#free-tier) models only. Models outside it return `402 model_requires_paid` until your first top-up.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.