xs83 · HackerTrans

9 months agodiscuss

If you mean CUDA specific then yes. The biggest benefit of these machines over the others is the CUDA ecosystem and tools like cuDF, cuGraph etc

xs83

9 months agodiscuss

TTFT on a Mac is terrible and only increases as the context increases, thats why many are selling their M3 Ultra 512GB

xs83

9 months agodiscuss

Now this looks much more interesting! Is the top one input tokens and the second one output tokens?

So 38.54 t/s on 120B? Have you tested filling the context too?

comments