MiniMax M1 40k: benchmark results
MiniMax M1 open hybrid-attention reasoning model at the 40K thinking budget. Provider: MiniMax. Released 2025-06-01. Access: Open.
Unified ELO 1560 ± 1, rank #403 of 1392 rated models, from 22 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| LLM Stats (OpenAI-MRCR: 2 needle 1M) | 58.6 | Score (%) | 100 |
| LLM Stats (OpenAI-MRCR: 2 needle 128k) | 76.1 | Score (%) | 87.5 |
| SpeechMap Compliance | 87.2 | % Requests Completed | 86.3 |
| LLM Stats (LongBench v2) | 61 | Score (%) | 62.5 |
| LLM Stats (AIME 2024) | 83.3 | Score (%) | 59.6 |
| BenchmarkList ECI | 115.34 | Capability Index (ECI) | 55 |
| ZeroEval MATH-500 | 96 | MATH-500 Score | 51.6 |
| AA Humanity's Last Exam | 7.83 | Accuracy (%) | 48.6 |
| LLM Stats Score | 20.81 | LLM Stats Score (conservative rating) | 48.4 |
| AA GPQA Diamond | 68.18 | Accuracy (%) | 45.4 |
| Artificial Analysis Intelligence Index | 8.5 | Intelligence Index | 44.3 |
| ZeroEval GPQA Diamond | 69.2 | GPQA Diamond Score | 41.7 |
Interactive version: theaggregate.ai/model?slug=minimax-m1-40k · How It Works · Data refreshed daily, snapshot 2026-09-05.