Ramp Launches SWE-Bench v2 as Top Models Converge in Performance
Ramp released SWE-Bench v2, a coding agent benchmark built on 500 real production tasks from its 200+ engineers. Astra topped the leaderboard at 76.4%, with top models converging in performance while differing sharply in cost.
2026-11-19 ~ 2026-11-19 · 2 related posts
- Ramp releases SWE-Bench v2: 500 production-built tasks, 6.2 Astra tops at 76.4% — AniBaddepudi · 2026-11-19
- Ramp SWE-Bench v2: top four coding models within points, but spend varies wildly — charliermarsh · 2026-11-19