7 ms·

I dont think this is Cerebras. Running on cerebras would change model behavior a bit and it could potentially get a ~10x speedup and it'd be more expensive. So

by diwank 8mo ago

I dont think this is Cerebras. Running on cerebras would change model behavior a bit and it could potentially get a ~10x speedup and it'd be more expensive. So most likely this is them writing new more optimized kernels for Blackwell series maybe?

simianwords 8mo ago
Fair point but it remains to answer - why isn’t this speed up available in ChatGPT and only in the api?