GeekRadar / The comparison desk
COMPARE.
AI models
minimax

MiniMax-M2.1 Preview

Model evaluation

Capability & preference

Human preference · Arena

Overall text preferences from head-to-head votes. Not factual accuracy or a task-success rate.

MiniMax-M2.1 Preview 1392.0 MiniMax-M2.1 Preview · 95% interval 1386.9–1397.1 · 17,222 votes · 2026-10-28

Recorded results

Only the same benchmark, version, workload, scenario and unit share a row. Missing results are not zero. System results include the full submitted hardware and software configuration.

Arena text preference / chinesehuman preference · 2026-10-28 · rating

MiniMax-M2.1 Preview

1,429.92 rating

Confidence interval: 1,414.14 – 1,445.71

1355 votes

Category
chinese
Evaluated configuration
MiniMax-M2.1 Preview
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-28 · rating

MiniMax-M2.1 Preview

1,420.07 rating

Confidence interval: 1,410.19 – 1,429.94

3475 votes

Category
coding
Evaluated configuration
MiniMax-M2.1 Preview
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-28 · rating

MiniMax-M2.1 Preview

1,362.00 rating

Confidence interval: 1,350.85 – 1,373.14

2806 votes

Category
creative_writing
Evaluated configuration
MiniMax-M2.1 Preview
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-28 · rating

MiniMax-M2.1 Preview

1,402.48 rating

Confidence interval: 1,395.02 – 1,409.93

6743 votes

Category
english
Evaluated configuration
MiniMax-M2.1 Preview
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating

MiniMax-M2.1 Preview

1,367.85 rating

Confidence interval: 1,360.69 – 1,375.01

12125 votes

Category
exclude_ties
Evaluated configuration
MiniMax-M2.1 Preview
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-28 · rating

MiniMax-M2.1 Preview

1,429.73 rating

Confidence interval: 1,412.83 – 1,446.64

1189 votes

Category
expert
Evaluated configuration
MiniMax-M2.1 Preview
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-28 · rating

MiniMax-M2.1 Preview

1,403.88 rating

Confidence interval: 1,375.88 – 1,431.88

446 votes

Category
french
Evaluated configuration
MiniMax-M2.1 Preview
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-28 · rating

MiniMax-M2.1 Preview

1,381.28 rating

Confidence interval: 1,346.12 – 1,416.45

297 votes

Category
german
Evaluated configuration
MiniMax-M2.1 Preview
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-28 · rating

MiniMax-M2.1 Preview

1,410.47 rating

Confidence interval: 1,403.91 – 1,417.03

9307 votes

Category
hard_prompts
Evaluated configuration
MiniMax-M2.1 Preview
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating

MiniMax-M2.1 Preview

1,418.07 rating

Confidence interval: 1,408.18 – 1,427.97

3558 votes

Category
hard_prompts_english
Evaluated configuration
MiniMax-M2.1 Preview
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating

MiniMax-M2.1 Preview

1,389.93 rating

Confidence interval: 1,379.49 – 1,400.37

3192 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
MiniMax-M2.1 Preview
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-28 · rating

MiniMax-M2.1 Preview

1,362.36 rating

Confidence interval: 1,352.15 – 1,372.57

3254 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
MiniMax-M2.1 Preview
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-28 · rating

MiniMax-M2.1 Preview

1,410.68 rating

Confidence interval: 1,394.13 – 1,427.22

1218 votes

Category
industry_legal_and_government
Evaluated configuration
MiniMax-M2.1 Preview
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-28 · rating

MiniMax-M2.1 Preview

1,416.26 rating

Confidence interval: 1,405.20 – 1,427.33

2759 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
MiniMax-M2.1 Preview
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-28 · rating

MiniMax-M2.1 Preview

1,418.94 rating

Confidence interval: 1,398.26 – 1,439.62

720 votes

Category
industry_mathematical
Evaluated configuration
MiniMax-M2.1 Preview
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-28 · rating

MiniMax-M2.1 Preview

1,386.43 rating

Confidence interval: 1,369.14 – 1,403.71

1183 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
MiniMax-M2.1 Preview
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-28 · rating

MiniMax-M2.1 Preview

1,415.33 rating

Confidence interval: 1,407.44 – 1,423.23

5760 votes

Category
industry_software_and_it_services
Evaluated configuration
MiniMax-M2.1 Preview
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-28 · rating

MiniMax-M2.1 Preview

1,373.02 rating

Confidence interval: 1,363.73 – 1,382.31

4038 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
MiniMax-M2.1 Preview
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-28 · rating

MiniMax-M2.1 Preview

1,400.06 rating

Confidence interval: 1,391.43 – 1,408.70

4654 votes

Category
instruction_following
Evaluated configuration
MiniMax-M2.1 Preview
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-28 · rating

MiniMax-M2.1 Preview

1,284.15 rating

Confidence interval: 1,237.41 – 1,330.89

177 votes

Category
japanese
Evaluated configuration
MiniMax-M2.1 Preview
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-10-28 · rating

MiniMax-M2.1 Preview

1,299.58 rating

Confidence interval: 1,264.23 – 1,334.93

285 votes

Category
korean
Evaluated configuration
MiniMax-M2.1 Preview
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-28 · rating

MiniMax-M2.1 Preview

1,415.36 rating

Confidence interval: 1,406.64 – 1,424.08

4540 votes

Category
longer_query
Evaluated configuration
MiniMax-M2.1 Preview
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-28 · rating

MiniMax-M2.1 Preview

1,396.47 rating

Confidence interval: 1,378.66 – 1,414.28

1044 votes

Category
math
Evaluated configuration
MiniMax-M2.1 Preview
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-28 · rating

MiniMax-M2.1 Preview

1,396.35 rating

Confidence interval: 1,385.27 – 1,407.42

2831 votes

Category
multi_turn
Evaluated configuration
MiniMax-M2.1 Preview
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-28 · rating

MiniMax-M2.1 Preview

1,377.68 rating

Confidence interval: 1,371.44 – 1,383.91

10479 votes

Category
non_english
Evaluated configuration
MiniMax-M2.1 Preview
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-28 · rating

MiniMax-M2.1 Preview

1,391.99 rating

Confidence interval: 1,386.92 – 1,397.07

17222 votes

Category
overall
Evaluated configuration
MiniMax-M2.1 Preview
Evaluation type
human preference
Arena text preference / polishhuman preference · 2026-10-28 · rating

MiniMax-M2.1 Preview

1,371.19 rating

Confidence interval: 1,346.30 – 1,396.09

529 votes

Category
polish
Evaluated configuration
MiniMax-M2.1 Preview
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-28 · rating

MiniMax-M2.1 Preview

1,386.67 rating

Confidence interval: 1,375.29 – 1,398.04

2387 votes

Category
russian
Evaluated configuration
MiniMax-M2.1 Preview
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-28 · rating

MiniMax-M2.1 Preview

1,396.68 rating

Confidence interval: 1,369.92 – 1,423.44

490 votes

Category
spanish
Evaluated configuration
MiniMax-M2.1 Preview
Evaluation type
human preference

Run it locally

Inference · memory estimate

MiniMax-M2.1 Preview

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license
MIT