On: Cognition's SWE-2 achieves 92.8 on Terminal-Bench 2.1

[walrus01]

Not really news, terminal bench 4 is the new metric. It's only a few points ahead in terminal bench 4 of some open weight models you can run on a 256GB system.

[ricardobeat → walrus01]

"Not really news" that a model from a smaller lab can beat weeks-old Fable 5.1 at 70% lower cost? What a time to be living in.

[p1esk → ricardobeat]

It’s very far from Fable on benchmarks that matter, like TB4.

[walrus01 → ricardobeat]

no, I mean not really news specifically on the number for terminal bench 2.

[llm_nerd → ricardobeat]

They did reinforcement learning on Kimi K3, probably specifically targeting the benchmarks.

Eh, it isn't news. I mean, it's just an echo bit of news to the great K3 release.