diff --git a/docs/customize/model-providers/more/lynkr.mdx b/docs/customize/model-providers/more/lynkr.mdx new file mode 100644 index 00000000000..acfdf314016 --- /dev/null +++ b/docs/customize/model-providers/more/lynkr.mdx @@ -0,0 +1,43 @@ +--- +title: "How to Configure Lynkr with Continue" +sidebarTitle: "Lynkr" +--- + + + Lynkr is an open-source, self-hosted LLM gateway that routes each request by + complexity: simple requests go to local models (Ollama, llama.cpp, LM Studio), + complex ones to a configured cloud provider (AWS Bedrock, Azure OpenAI, + Databricks, OpenRouter, and others). It also strips unused tool schemas and + compresses large JSON tool results to reduce token usage. + + + + Get started with [Lynkr on GitHub](https://github.com/Fast-Editor/Lynkr) + + +## Installation + +Lynkr runs locally and provides an OpenAI-compatible API: + +```bash +npm install -g lynkr +lynkr init # interactive wizard: choose tier models and provider credentials +lynkr start +``` + +This starts the gateway at `http://localhost:8081`. + +## Configuration + +```yaml title="config.yaml" +models: + - name: Lynkr + provider: openai + model: lynkr-auto + apiBase: http://localhost:8081/v1 + apiKey: none +``` + +The `model` value is a placeholder — Lynkr's tier routing selects the actual +model per request based on the tiers configured in `lynkr init`, so a single +entry covers the local-to-cloud range. diff --git a/docs/docs.json b/docs/docs.json index b7a1d83f13a..978845ec493 100644 --- a/docs/docs.json +++ b/docs/docs.json @@ -112,6 +112,7 @@ "customize/model-providers/more/groq", "customize/model-providers/more/llamacpp", "customize/model-providers/more/llamastack", + "customize/model-providers/more/lynkr", "customize/model-providers/more/mimo", "customize/model-providers/more/mistral", "customize/model-providers/more/moonshot",