MiniMaxMiniMax M2
MiniMax M2.7
MiniMax · MiniMax M2. Released Mar 18, 2026.
12 results
Benchmark results 12
Compare this model| Benchmark | Category | Score | Metric | Status | Date | |
|---|---|---|---|---|---|---|
| SWE-bench Pro | Coding | 56.22% | % resolved | Verified | Mar 18, 2026 | Details |
| SWE-bench Verified | Coding | 78% | % resolved | Unverified | Mar 18, 2026 | Details |
| Artificial Analysis Intelligence Index | Reasoning | 49.6 | Index | Unverified | Mar 18, 2026 | Details |
| SWE-rebench | Coding | 51.9% | Resolved rate (pass@1) | Unverified | Mar 18, 2026 | Details |
| τ²-bench (Telecom) | Tool use | 84.8% | pass^1 | Official | Mar 18, 2026 | Details |
| GDPval | Agents | 1505 | Elo | Official | Mar 18, 2026 | Details |
| SciCode | Coding | 47.0% | accuracy | Unverified | Mar 18, 2026 | Details |
| AA-Omniscience: Knowledge and Hallucination Benchmark | Reasoning | 1 | AA-Omniscience Index | Official | Mar 18, 2026 | Details |
| IFBench | Reasoning | 75.7% | accuracy | Official | Mar 18, 2026 | Details |
| WebDev Arena | Chat preference | 1401 | Elo | Verified | Mar 18, 2026 | Details |
| Design Arena | Chat preference | 1285 | Elo | Verified | Mar 18, 2026 | Details |
| Vectara Hallucination Leaderboard | Other | 12.9% | Hallucination Rate | Official | Mar 18, 2026 | Details |