Poolside for GitHub Copilot Chat

Use hosted Poolside coding models directly from the GitHub Copilot Chat model picker in Visual Studio Code.

This extension is a native VS Code LanguageModelChatProvider. It validates a user-supplied Poolside Platform API key, discovers the hosted models available to that key, and streams responses from inference.poolside.ai into Copilot Chat without a local proxy.

Highlights

• Direct, first-class Poolside Platform integration
• Credentials managed by VS Code native secret provider configuration
• Multiple Poolside API-key entries in VS Code's Manage Language Models flow
• Live hosted-model discovery with sensible Laguna fallbacks
• Streaming text and model reasoning
• Configurable reasoning effort in the Copilot model picker (max for thinking on, none for off)
• Agent mode function-tool calls
• Native VS Code context-window accounting from Poolside usage data
• No proxy, bundled server, or third-party relay

Quick start

1. Install Poolside for GitHub Copilot Chat. You need VS Code 1.125 or newer and GitHub Copilot Chat.
2. Create a developer API key in Poolside Platform.
3. Open Copilot Chat, select Manage Models, add a Poolside provider entry, name it (for example work), and enter your API key. VS Code stores the provider credential securely and Poolside model discovery checks it when loading models.
4. Choose an available Laguna model.

To use more than one Poolside account or API key, add another Poolside entry in Manage Language Models with a different name and its API key. Each entry's models appear as their own group in the model picker; rename an entry any time without affecting stored keys.

Use Poolside: Manage Connection to test the connection, refresh hosted models, set the inline suggestions model, inspect logs, or open diagnostics.

Choose Max or None from the thinking control in Copilot Chat. Poolside-hosted inference defaults to Max; the selection applies to that request and overrides the poolsideCopilot.reasoningEffort workspace default.

Poolside's hosted Laguna models are currently text-only. Image attachments are disabled; prompts, tool definitions, tool results, and conversation context selected by Copilot Chat are sent directly to Poolside for inference.

Group-name-native entries

Entries need only an API key. The name you give an entry in Manage Language Models is its identity: VS Code namespaces each entry's models per group, keeps credentials in its secure storage, and offers key updates on the entry. Rotating a key updates the entry's credential scope automatically; no entry IDs, alias blocks, or manual restore steps remain. Deleting an entry removes its VS Code-owned credential.

Upgrading from earlier releases

Recreate existing entries in Manage Language Models with the new single-field flow: each entry now needs only a name and its API key — explicit entry IDs, selection commands, and forgotten-entry blocks no longer exist. Select your Laguna models again after re-adding entries; deleting an entry removes its VS Code-owned credential.

Context token counts are estimates. Tool-result text, tool arguments, thinking, and image inputs are included in context budgeting. If the service rejects a full context window, run /compact in Copilot Chat, reduce attachments or tool output, or start a new chat.

Documentation

• Setup, settings, and troubleshooting
• Models and pricing
• API key and security model
• Development and releases

Related projects

• Grok for GitHub Copilot Chat — Use xAI Grok models directly from the GitHub Copilot Chat model picker.
• Codex Bridge for Copilot Chat — Use OpenAI Codex models in Copilot Chat with a ChatGPT Plus or Pro subscription.
• Ollama Cloud for GitHub Copilot Chat — Use Ollama Cloud models with native thinking and tool support.
• OpenCode for GitHub Copilot Chat — Use OpenCode Zen, Go, and Console models from the model picker.

Unofficial project; not affiliated with Poolside, GitHub, or Microsoft. Poolside account limits and charges still apply. Licensed under MIT.