https://cognition.ai/blog/frontier-code (disclaimer - was on the team - but also... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		swyx 3 days ago \| parent \| context \| favorite \| on: Separating signal from noise in coding evaluations https://cognition.ai/blog/frontier-code (disclaimer - was on the team - but also we covered swebench pro/deepswe issues in here as well.)
		help

sk4rekr0w 2 days ago [–]

1.1 seems a lot better than the original release, which was a bit hyperbolic. excited to see the team keep iterating.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact