yudaprasetya007/routeVSCODE

⚡ Zero-reload dynamic AI model switcher & local proxy for VSCode Copilot Chat via 9Router

What it solves

GitHub Copilot Chat in VSCode typically requires manual configuration changes or window reloads to switch between different AI models. This project provides a way to switch models (such as Claude 3.5 Sonnet, GPT-4o, Gemini 1.5 Pro, and DeepSeek) instantly without restarting the editor or interrupting chat sessions.

How it works

The tool uses a "Zero-Reload Dynamic Local Proxy" (running on port 20129) that sits between VSCode and the 9Router AI gateway. When a user selects a new model via a web dashboard or the VSCode status bar, the proxy updates the active model in memory. When Copilot Chat sends a request, the proxy intercepts the payload, injects the currently active model ID, and forwards it to the 9Router core engine, which then routes the request to the appropriate AI provider.

Who it’s for

Developers using GitHub Copilot Chat in Visual Studio Code who want to quickly experiment with and compare different LLMs from various providers (OpenAI, Anthropic, Google, etc.) without configuration friction.

Highlights

  • Zero-Reload Switching: Change AI models instantly without restarting VSCode.
  • Web Dashboard: A visual interface for model selection and live model comparison/benchmarking.
  • Status Bar Integration: Quick model switching directly from the VSCode UI.
  • Broad Provider Support: Connects to over 40 AI providers via the 9Router gateway.
  • Dynamic Local Proxy: Automatically handles model injection, SSE token streaming, and authentication headers.

Related

  • Project
    Trianglezichopper/submagic-core

    Submagic Core is a lightweight AI environment for Windows that enables model chaining, workflow automation, and side-by-side model comparison.

  • Project
    Linzwcs/EvoPolicyGym

    An evaluation framework that measures how coding agents evolve executable policies through iterative experimentation and feedback in interactive environments.

  • Project
    yureii1996/cek-probe-model

    A collection of Python scripts to test OpenAI-compatible model endpoints to verify if the model being served matches its label.

  • Project
    OctaFuse/octafuse-gateway

    Octafuse Gateway is an open‑source, self‑hosted AI gateway that unifies access to multiple model providers (OpenAI, Anthropic, Gemini, DashScope, etc.) and agent tools behind a single OpenAI‑compatible API. It offers one‑click provider/model import, protocol translation, four routing strategies, detailed user/quota/billing management, an admin UI, and can be deployed via Docker or Cloudflare Workers.