Skip to main content
VS Code’s Copilot chat can use your own model provider (Bring Your Own Key) through its Custom Endpoint provider. Point it at Kyma’s Chat Completions endpoint to use Kyma models in chat and agent mode.

​
Configuration

  1. In the chat model picker, select Manage Language Models (gear icon), or run Chat: Manage Language Models from the Command Palette.
  2. Select Add Models → Custom Endpoint, enter a group name (for example Kyma) and your Kyma API key, and choose the Chat Completions API type.
  3. VS Code opens chatLanguageModels.json. Add the models you want:
  1. Save the file and pick the model from the chat model picker. Restart VS Code if it does not appear.
Each id must be an id from GET /v1/models. Agent mode only lists models with "toolCalling": true. A new account’s signup credit is added once your email is verified and spends on free tier models only. Models outside it return 402 model_requires_paid until your first top-up. See all model aliases for shortcuts.

​
Notes

  • Bring Your Own Key covers Copilot chat and utility tasks. Inline suggestions (code completions) still use GitHub’s own models, so Kyma does not serve them.
  • Copilot Business and Enterprise admins can turn Bring Your Own Key off by policy.

​
Prompt Caching

Some models cache a repeated prompt prefix. pricing.cached on /v1/models is the rate Kyma publishes for cached input on that model — usually below its input rate, but not always, so compare the two for the model you use. Every chat completion reports usage.cost, what the request was charged, and usage.cache_discount, Kyma’s figure for what caching saved on it. Whether a prefix is cached automatically or needs a cache_control marker depends on the model’s creator. See Prompt Caching for details.