Follow
Llama 3.1 405B now runs at 969 tokens/s on Cerebras Inference
https://news.ycombinator.com/item?id=42178761
#hackernews #tech
