AARRI-Bench (Mini-SWE-Agent) - Hands-on: leaderboard

Metric: Task success rate (%) under the 0/1 reward on the hands-on tasks: coding, experimental setup and data processing; Harbor containerized runs, Mini-SWE-Agent agent harness; a configuration's tasks without a valid run are left out of its denominator; higher is better. Source: arxiv.org. Saturation forecast: Around December 2026. 8 models tracked.

Top models

#ModelScore
1Qwen 3.6 Plus64.3
2Claude Opus 4.757.1
3Kimi K2.650
4DeepSeek V4 Flash50
5MiniMax-M2.742.9
6Qwen 3 235B A22B 2507 (Thinking)35.7
7GPT-OSS-120B30.8
8Qwen 3 Next 80B A3B Instruct28.6

Interactive version: theaggregate.ai/benchmark?slug=aarri-bench-mini-swe-agent-hands-on · How It Works · Data refreshed daily, snapshot 2026-09-29.