GitHub Copilot with ShareQuota

VS Code’s Copilot Chat can register custom OpenAI-compatible models through the customOAIModels setting: a name, a URL, and what the model supports. Point the URL at the local endpoint, choose it in the model picker, and paste your sq- key when prompted. Chat then runs against the plan you connected, while Copilot’s own completions stay on Copilot.

Setup

1. 1Start the proxy and connect a planRun the binary, open the dashboard, and sign in to the accounts you want Copilot Chat to use.
2. 2Register the modelAdd the customOAIModels entry to VS Code settings.json with the local /v1 URL and a routed model id as the key.
3. 3Pick it in the model pickerOpen Copilot Chat, choose Manage Models, select the entry and paste your sq- key when asked.

What to set

{  "github.copilot.chat.customOAIModels": {    "claude/claude-sonnet-5": {      "name": "Claude Sonnet 5 via ShareQuota",      "url": "http://localhost:20130/v1",      "toolCalling": true,      "vision": false,      "maxInputTokens": 200000,      "maxOutputTokens": 8192    }  }}

Worth knowing

• Bring-your-own-key applies to Copilot Chat in VS Code; inline completions keep using Copilot’s own models.
• Custom models are a VS Code feature that has changed name and shape more than once; the docs link above is the current source if the setting key differs.

To undo it

Remove the entries above, or turn the integration off in the dashboard and the backup is restored. Nothing else was written.

Questions

Checked against the tool’s own documentation on 2026-09-13. Settings move; if a field has been renamed, the docs link above is the source.