AutoRLBench 38 Issue / 19 Solution - Solution: leaderboard

Metric: Repaired run passes its rerun, 19 tasks (%). Source: thinking-space.github.io. Saturation forecast: Around September 2028. 9 models tracked. Updated 18 Nov 2026.

Top models

#ModelSettingScoreOverall rank
1GPT-6.1 SolMax42.1#12
2GPT-6 AstraMax36.8#7
3GLM-5.3Max36.8#72
4Kimi K3.1 ProMax36.8#32
5Qwen 3.8 MaxMax31.6#63
6Claude Fable 5.1Max26.3#17
7Kimi K3Max21.1#60
8Grok 4.7Max21.1#48
9DeepSeek V4 ProMax10.5#194

No result here: #1 Claude Fable 5.2, #2 GPT-6.2 Astra, #3 Gemini 4 Carbon.

Interactive version: theaggregate.ai/benchmark?slug=autorlbench-38-issue-19-solution-solution · How It Works · Data refreshed daily, snapshot 2026-11-18.