Claw-SWE-Bench (nanobot): leaderboard

Metric: Pass@1 (%) on the 350 GitHub issue-resolution instances of Claw-SWE-Bench (8 languages, 43 repositories, from SWE-bench-Multilingual and SWE-bench-Verified-Mini after future-commit cleanup); full adapter, identical prompt, 60-minute per-instance budget, one run, SWE-bench evaluator; nanobot harness. Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 2 models tracked.

Top models

#ModelScore
1nanobot + GLM-5.160.9
2nanobot + Qwen3.6-Flash47.4

Interactive version: theaggregate.ai/benchmark?slug=claw-swe-bench-nanobot · How It Works · Data refreshed daily, snapshot 2026-09-29.