Qwen Remote Assistant & GPU Telemetry

Connects VS Code to a remote self-hosted Ollama server running on an NVIDIA RTX GPU via Cloudflare Tunnel or Tailscale.

Features

• Live GPU Telemetry: Displays GPU temperature, VRAM usage, and core load in the status bar.
• Encrypted Token Storage: Stores your secret token using Windows Credential Manager / VS Code SecretStorage API.
• Code Generation: Select code and stream generation directly from your remote Qwen model.

Setup

1. Run Qwen: Set Secret Auth Token from the Command Palette (Ctrl+Shift+P).
2. Set your Cloudflare Tunnel URL or Tailscale address under Settings -> Qwen Remote AI.