MMLU-Pro#6 of 28
exact_match
View run →75.6%±0.4%
Benchmark results loaded.
MiniMax/MiniMax-M2.1-AWQ
- Weights
- 124.9 GB · safetensors
- Source
- huggingface page
- Commit
- 0909e6d
Evaluated across 12 benchmarks. Ranks in the top 3 on 0 of 12. Strongest showing on MMLU-Pro (75.6% exact_match, #6 of 28). Weakest on MBPP(Instruct) (0.0% pass@1, #22 of 24).
Benchmark results
| Benchmark | Metric | Score | Rank | Actions |
|---|---|---|---|---|
| MMLU-Pro | exact_match | 75.6%±0.4% | #6 of 28 | View run → |
| GSM8K | exact_match | 70.4%±1.3% | #18 of 28 | View run → |
| MBPP | pass@1 | 60.4%±2.2% | #10 of 28 | View run → |
| EQ-Bench | eqbench | 52.5±3.1 | #24 of 28 | View run → |
| MGSM | exact_match | 48.2%±0.8% | #25 of 28 | View run → |
| IFEval | prompt_level_strict_acc | 35.3%±2.1% | #27 of 28 | View run → |
| GPQA Diamond | acc | 28.8%±3.2% | #17 of 28 | View run → |
| MMLU | acc | 25.5%±0.4% | #27 of 28 | View run → |
| GPQA Main | acc | 24.3%±2.0% | #26 of 28 | View run → |
| GPQA Extended | acc | 23.4%±1.8% | #27 of 28 | View run → |
| BBH | exact_match | 0.1%±0.0% | #25 of 28 | View run → |
| MBPP(Instruct) | pass@1 | 0.0%±0.0% | #22 of 24 | View run → |
- GSM8K#18 of 28exact_matchView run →70.4%±1.3%
- MBPP#10 of 28pass@1View run →60.4%±2.2%
- EQ-Bench#24 of 28eqbenchView run →52.5±3.1
- MGSM#25 of 28exact_matchView run →48.2%±0.8%
- IFEval#27 of 28prompt_level_strict_accView run →35.3%±2.1%
- GPQA Diamond#17 of 28accView run →28.8%±3.2%
- MMLU#27 of 28accView run →25.5%±0.4%
- GPQA Main#26 of 28accView run →24.3%±2.0%
- GPQA Extended#27 of 28accView run →23.4%±1.8%
- BBH#25 of 28exact_matchView run →0.1%±0.0%
- MBPP(Instruct)#22 of 24pass@1View run →0.0%±0.0%
last evaluated:1 weeks ago
How to cite
Citation
FrozeBench. "MiniMax/MiniMax-M2.1-AWQ." https://frozebench.com/models/minimax-minimax-m2-1-awq. Retrieved 2026-06-04.
BibTeX
@misc{frozebench_MiniMax_MiniMax_M2_1_AWQ,
title = {MiniMax/MiniMax-M2.1-AWQ},
howpublished = {\url{https://frozebench.com/models/minimax-minimax-m2-1-awq}},
year = {2026},
note = {FrozeBench. Retrieved 2026-06-04.}
}URL
https://frozebench.com/models/minimax-minimax-m2-1-awq