PinchBench - 147 Tasks - OpenClaw 2026.5.20 (e510042): leaderboard

Metric: Best Official Run Score (%). Source: pinchbench.com. Saturation forecast: Not forecast. 39 models tracked.

Top models

#ModelScore
1MiMo-V2.591.87
2DeepSeek V4 Flash91.46
3Grok 4.591.17
4Claude Opus 4.791.11
5GPT-5.6 Luna90.79
6Claude Haiku 4.588.26
7Claude Opus 4.687.97
8Grok 4.2087.78
9GPT-5.6 Sol87
10MiMo-V2.5-Pro86.25
11GLM-5V Turbo85.33
12Grok 4.385.01
13GLM-5-Turbo84.69
14Qwen 3.6 Plus84.65
15MiniMax-M2.784.46

Interactive version: theaggregate.ai/benchmark?slug=pinchbench-147-tasks-openclaw-2026-5-20-e510042 · How It Works · Data refreshed daily, snapshot 2026-10-09.