alt Hacker NewsI really hope one day there is something like Opus 4.8 but with Cerebras' speed -- they reach over 1,000t/s on gpt-oss-120b but that model is seemingly not even properly trained for tool calling. But watching it slam out several entire screens of thinking/reasoning per second is amazing. I'd love that with Opus quality.
I like gpt oss - great model even if not too smart.. runs on my laptop at over 100ts has a certain tone that I like over all these qwens stuck up their asses.