Hackernews posts about 2 M2 Ultras

1. DeepSeek R1 671B running on 2 M2 Ultras faster than reading speed (twitter.com)
   96 points by thyrox 5 months ago | 29 comments
2. M2 Ultra can run 128 streams of Llama 2 7B in parallel (github.com)
   268 points by behnamoh over 1 year ago | 173 comments
3. Ask HN: Running LLMs locally–what hardware should I get–M2 Ultra, PC, cloud GPU?
   2 points by behnamoh over 1 year ago | 2 comments
4. Qcom elite uses per-cluster DVFS, with cores in a cluster share the same freq
   2 points by firstlovelife 7 months ago | discuss