2 ms·Demo: Run LLava multimodal LLM with a 2M Rust app portable across GPU/CPU2 points by 3Sophons 3y ago