Hacker Newsnew | past | comments | ask | show | jobs | submit | fromlogin
DeepSWE results are unreliable – 3/3 DSv4 "failed" tasks solved with same model (github.com/datacurve-ai)
3 points by theanonymousone 1 day ago | past | discuss
DeepSWE Audit: DeepSeek-v4-pro results are unreliable (github.com/datacurve-ai)
3 points by eunos 2 days ago | past | discuss

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

Search: