SWE-bench Science launches: best agent (Claude Code + Opus-5) under 50% pass@1
_akhaliq · x · 2026-08-21
A new benchmark, SWE-bench Science, targets scientific software engineering with 119 tasks across 98 repositories and 20 domains. Even the strongest agent—Claude Code with Opus-5—achieves under 50% pass@1, showing scientific computing code remains a clear weak spot for current coding agents.
Related event: SWE-bench Science Launches; Top Agents Pass Under 50%(3 posts)→
More from coding & agent
- How to hook Claude Code or Codex up to iMessage in 3 steps — jacob_posel · 2026-10-16
- Are the fastest GPU kernels now written by AI? Kernels are hyper-verifiable, argues one dev — fleetwood___ · 2026-10-16
- Anthropic lead engineer's 4-question prompt makes agents more ambitious, now a free skill — daniel_mac8 · 2026-10-16
- ChainForge dev shows 'feature map' for orchestrating subagents on large PRs — IanArawjo · 2026-10-16
- aiDotEngineer benchmarks agent orchestration pipelines, staging humans as Foreman and workers — vikvang1 · 2026-10-16
- Developer mass-produces 200 videos in hours using Claude Code + Hyperframes — mcraddock · 2026-10-16