tinygrad Posts Local Qwen Chat Setup Command
Official tweet shares command to run local Qwen model server with browser chat.
TLDR
The official @tinygrad account posted instructions to run python3 -m tinygrad.llm --model qwen3.8:27b --serve --max_context 65536 on a 7900XTX GPU. Users can then visit localhost:8000 for a browser-based chat interface titled tinygrad chat at teeny:8000. The post notes an added markdown parser and OpenAI API compatibility with pi and opencode. An attached screenshot shows the browser window with the chat setup.
Combined views
15.5K
1 Source, first seen 24d ago