Models / MiniMax M2.5

MiniMax

MiniMax M2.5

18 published results from 4 sources. Each card shows where the number comes from and what it does not measure. Results are never combined into one score.

Provider
MiniMax
Sources
4
Our benchmarks
0
Price
Not yet published

Reported by others

Reported by Arena (formerly LMArena)
1,386
Overall · rank 149 of 177
Unit
Arena rating, higher is better
Range
1,382 to 1,389
Sample
42398 votes
Configuration
MiniMax M2.5
Measured
25 Sep 2026
Not shown
Not an error rate: claims that cannot be checked on the web are skipped, and preference still carries most of the weight.
Reported by Arena (formerly LMArena)
1,400
Business, management and finance · rank 122 of 402
Unit
Arena rating, higher is better
Range
1,393 to 1,407
Sample
8377 votes
Configuration
MiniMax M2.5
Measured
25 Sep 2026
Not shown
Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
Reported by Arena (formerly LMArena)
1,357
Creative writing · rank 142 of 407
Unit
Arena rating, higher is better
Range
1,349 to 1,365
Sample
6640 votes
Configuration
MiniMax M2.5
Measured
25 Sep 2026
Not shown
Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
Reported by Arena (formerly LMArena)
1,425
Expert prompts · rank 113 of 359
Unit
Arena rating, higher is better
Range
1,414 to 1,435
Sample
3770 votes
Configuration
MiniMax M2.5
Measured
25 Sep 2026
Not shown
Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
Reported by Arena (formerly LMArena)
1,381
Instruction following · rank 146 of 409
Unit
Arena rating, higher is better
Range
1,375 to 1,387
Sample
13752 votes
Configuration
MiniMax M2.5
Measured
25 Sep 2026
Not shown
Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
Reported by Arena (formerly LMArena)
1,391
Overall · rank 156 of 409
Unit
Arena rating, higher is better
Range
1,387 to 1,395
Sample
42743 votes
Configuration
MiniMax M2.5
Measured
25 Sep 2026
Not shown
Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
Reported by Arena (formerly LMArena)
1,369
Writing, literature and language · rank 146 of 408
Unit
Arena rating, higher is better
Range
1,363 to 1,376
Sample
10049 votes
Configuration
MiniMax M2.5
Measured
25 Sep 2026
Not shown
Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
Reported by OpenHands Index
45.2%
Average · rank 22 of 34
Unit
% resolved, higher is better
Configuration
minimax-m2.5
Measured
30 Jun 2026
Not shown
Not the model alone: the score is for the model inside the OpenHands agent, at the SDK version shown.
Reported by OpenHands Index
25.0%
Front end (SWE-Bench Multimodal) · rank 27 of 34
Unit
% resolved, higher is better
Configuration
minimax-m2.5
Measured
30 Jun 2026
Not shown
Not the model alone: the score is for the model inside the OpenHands agent, at the SDK version shown.
Reported by OpenHands Index
12.5%
Greenfield (Commit0) · rank 25 of 34
Unit
% resolved, higher is better
Configuration
minimax-m2.5
Measured
30 Jun 2026
Not shown
Not the model alone: the score is for the model inside the OpenHands agent, at the SDK version shown.
Reported by OpenHands Index
47.9%
Information gathering (GAIA) · rank 24 of 34
Unit
% resolved, higher is better
Configuration
minimax-m2.5
Measured
30 Jun 2026
Not shown
Not the model alone: the score is for the model inside the OpenHands agent, at the SDK version shown.
Reported by OpenHands Index
72.6%
Issue resolution (SWE-Bench) · rank 23 of 34
Unit
% resolved, higher is better
Configuration
minimax-m2.5
Measured
30 Jun 2026
Not shown
Not the model alone: the score is for the model inside the OpenHands agent, at the SDK version shown.
Reported by OpenHands Index
68.1%
Testing (SWT-Bench) · rank 18 of 34
Unit
% resolved, higher is better
Configuration
minimax-m2.5
Measured
30 Jun 2026
Not shown
Not the model alone: the score is for the model inside the OpenHands agent, at the SDK version shown.
Reported by UGI Leaderboard
41.0%
Requested-length error · rank 294 of 370
Unit
% off the requested word count, lower is better
Configuration
MiniMax M2.5
Measured
12 Feb 2026
Not shown
Not other format limits such as character counts or bullet counts.
Reported by UGI Leaderboard
0.38
Style adherence · rank 71 of 370
Unit
score from 0 to 1, higher is better
Configuration
MiniMax M2.5
Measured
12 Feb 2026
Not shown
Not brand voice on your own examples: UGI's prompts are private and lean towards creative writing.
Reported by UGI Leaderboard
41.7
Writing score · rank 200 of 370
Unit
score out of 100, higher is better
Configuration
MiniMax M2.5
Measured
12 Feb 2026
Not shown
Not business copy quality: UGI's prompts are private and lean towards creative writing, and models that often refuse get no score.
Reported by Vectara Hallucination Leaderboard
98.2%
Answer rate · rank 80 of 108
Unit
% of documents, higher is better
Configuration
MiniMax M2.5
Measured
22 Sep 2026
Not shown
Not a quality score: a low rate usually means content filters were triggered, and hallucination rates are measured on answered documents only.
Reported by Vectara Hallucination Leaderboard
9.1%
Hallucination rate · rank 46 of 108
Unit
% of summaries, lower is better
Configuration
MiniMax M2.5
Measured
22 Sep 2026
Not shown
Not errors in open questions or other tasks: only summarisation, judged by Vectara's own model (HHEM-2.3), not by people, on news-style documents rather than your data.

Compare MiniMax M2.5 with