Llama 3.1 405B now runs at 969 tokens/s on Cerebras Inference

Cerebras Launches the Fastest AI Inference

Cerebras Inference: AI at Instant Speed