GitHub Copilot - Kyma API

GitHub Copilot supports custom OpenAI-compatible endpoints. You can route Copilot through Kyma to access Kyma’s active open models.

Configuration

In VS Code, open Settings (Cmd+,) and search for github.copilot:

Set Copilot: Advanced → API URL to:

https://kymaapi.com/v1

Set your Kyma API key in the authentication field.

Alternatively, set environment variables:

export OPENAI_API_KEY="ky-your-api-key"
export OPENAI_BASE_URL="https://kymaapi.com/v1"

Recommended Models

Model	Alias	Best For
`qwen-3-32b`	`fast`	Recommended — fast code completion
`qwen-3.6-plus`	`best`	Highest quality
`qwen-3-coder`	`code`	Best alias for code-focused tasks
`kimi-k2.6`	`agent`	Best tool calling
`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.

See all model aliases for shortcuts.

Notes

Kyma is OpenAI-compatible, so any Copilot setup that supports custom OpenAI endpoints will work.
For inline completions, use a fast model like qwen-3-32b or alias fast.

Prompt Caching

Prompt caching is automatic. Your system prompt and tool definitions are cached for 5 minutes, and the repeated part then bills at that model’s cached-input rate instead of its input rate. No code changes needed. See Prompt Caching for details.