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

cdnsteve 60 points 27 comments September 10, 2026
tokenstead.ai · View on Hacker News

Discussion Highlights (6 comments)

varispeed

These tests are pointless when often models get nerfed few days after release. Astra today is way dumber than just few days ago.

forgot-my-pw

Terminal Bench 2/2.1 appears to be almost solved, so we probably shouldn't look too hard on that? Their Terminal Bench 4 score is soso. Still quite impressive though.

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.

ChrisArchitect

Related: Cognition launches new SWE-2 model https://news.ycombinator.com/item?id=49645443

captainregex

my personal experience with swe has been…suboptimal. I am not sure how much I buy these benchmarks and it has a very “just blurt it out even if it’s probably not right” style but hey it’s free.

samusiam

Which is pretty much a useless (i.e., saturated, contaminated) benchmark now.

Semantic search powered by Rivestack pgvector
6,164 stories · 56,060 chunks indexed