A fast inference library for running LLMs locally on modern consumer-class GPUs *with support for qwen (untested)*
By chatting or signing in you agree to the Terms and chat-message logging (revocable in History).