MMPCBench - Strategic Effectiveness: leaderboard
Metric: Judge-panel score (0-2; detected errors only). Source: arxiv.org. 14 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.1 (High) | 1.95 |
| 2 | GPT-5 Mini | 1.93 |
| 3 | GPT-5.2 | 1.87 |
| 4 | Claude Sonnet 4.5 (Thinking) | 1.86 |
| 5 | Gemini 3 Pro | 1.79 |
| 6 | Gemini 3 Flash | 1.77 |
| 7 | Qwen 3 VL 235B A22B Instruct | 1.69 |
| 8 | GLM-4.6V | 1.69 |
| 9 | Qwen 3 VL 30B A3B Instruct | 1.57 |
| 10 | Qwen 3 VL 8B Instruct | 1.54 |
| 11 | Gemma 3 27B (IT) | 1.42 |
| 12 | Gemma 3 12B (IT) | 1.42 |
| 13 | Gemma 3 4B (IT) | 1.24 |
Interactive version: theaggregate.ai/benchmark?slug=mmpcbench-strategic-effectiveness · How It Works · Data refreshed daily, snapshot 2026-09-19.