3 ms·
Running LLaMA-70B in the browser with MLC
- swyx 3y ago(cohost here) this was a fantastic conversation about the potential of machine learning compilation to run truly large models on phones and browsers. really excited to share it with the HN community, please let us know if there's any followups or questions!