https://taalas.com/
Is an example startup in this area claiming 16k tok/s on an asic for llama 8b. Qwen has a 27b model at opus 4.5 quality.
https://taalas.com/
Is an example startup in this area claiming 16k tok/s on an asic for llama 8b. Qwen has a 27b model at opus 4.5 quality.