Comment on DeepSeek-V3 now runs at 20 tokens per second on Mac Studio, and that’s a nightmare for OpenAI
morrowind@lemmy.ml 6 days agoThe he’ll is v3 32b. Are you talking about a distill
Comment on DeepSeek-V3 now runs at 20 tokens per second on Mac Studio, and that’s a nightmare for OpenAI
morrowind@lemmy.ml 6 days agoThe he’ll is v3 32b. Are you talking about a distill
vintageballs@feddit.org 1 day ago
They probably confused the R1 Qwen distill with something else. Afaik there is no 32b model from DeepSeek directly.