Augment Code @augmentcode.com · Mar 31

🧵 We just released the #1 open-source agent on the SWE-bench Verified leaderboard by assembling the best of Claude Sonnet 3.7 and O1. Open-source repo here: github.com/augmentcode/... Here's how we achieved 65.4% success rate on the hardest coding benchmark in the industry: 🧠👇

2 likes 1 replies

?

Replies

Augment Code · Mar 31

At Augment, our mission is building the best AI platform for professional software engineers. We obsessively test, fine-tune, and optimize models so teams can harness AI without tracking which model does what today (when it will all change again tomorrow).