MiniMaxMiniMax M2
MiniMax M2.5
MiniMax · MiniMax M2. Released Feb 12, 2026.
13 results
Benchmark results 13
Compare this model| Benchmark | Category | Score | Metric | Status | Date | |
|---|---|---|---|---|---|---|
| SWE-bench Verified | Coding | 80.2% | % resolved | Verified | Feb 12, 2026 | Details |
| SWE-bench Pro | Coding | 55.4% | % resolved | Verified | Feb 12, 2026 | Details |
| GPQA Diamond | Reasoning | 85.2% | accuracy | Verified | Feb 12, 2026 | Details |
| Humanity's Last Exam | Reasoning | 19.4% | accuracy | Verified | Feb 12, 2026 | Details |
| Epoch Capabilities Index | Reasoning | 147.4 | Index | Official | Feb 12, 2026 | Details |
| GDPval | Agents | 1176 | Elo | Official | Feb 12, 2026 | Details |
| SciCode | Coding | 42.6% | accuracy | Unverified | Feb 12, 2026 | Details |
| Global-MMLU | Reasoning | 84.2% | accuracy | Unverified | Feb 12, 2026 | Details |
| WebDev Arena | Chat preference | 1382 | Elo | Verified | Feb 12, 2026 | Details |
| EQ-Bench Creative Writing v3 | Chat preference | 1331 | Elo | Verified | Feb 12, 2026 | Details |
| Design Arena | Chat preference | 1261 | Elo | Verified | Feb 12, 2026 | Details |
| SWE-bench Multilingual | Coding | 68.3% | % resolved | Official | Feb 12, 2026 | Details |
| Vectara Hallucination Leaderboard | Other | 9.1% | Hallucination Rate | Official | Feb 12, 2026 | Details |