GeekRadar / The comparison desk
COMPARE.
AI models
mistral

Mistral Medium 3 (2505)

Model evaluation

Capability & preference

Human preference · Arena

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

Mistral Medium 3 (2505) 1370.1 Mistral Medium 3 (2505) · 95% interval 1365.4–1374.8 · 32,777 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

Mistral Medium 3 (2505)

1,373.52 rating

Confidence interval: 1,360.57 – 1,386.46

2047 votes

Category
chinese
Evaluated configuration
Mistral Medium 3 (2505)
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-28 · rating

Mistral Medium 3 (2505)

1,385.50 rating

Confidence interval: 1,377.51 – 1,393.50

5842 votes

Category
coding
Evaluated configuration
Mistral Medium 3 (2505)
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-28 · rating

Mistral Medium 3 (2505)

1,343.38 rating

Confidence interval: 1,333.89 – 1,352.87

4063 votes

Category
creative_writing
Evaluated configuration
Mistral Medium 3 (2505)
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-28 · rating

Mistral Medium 3 (2505)

1,384.47 rating

Confidence interval: 1,378.57 – 1,390.37

15108 votes

Category
english
Evaluated configuration
Mistral Medium 3 (2505)
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating

Mistral Medium 3 (2505)

1,340.28 rating

Confidence interval: 1,333.90 – 1,346.65

22753 votes

Category
exclude_ties
Evaluated configuration
Mistral Medium 3 (2505)
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-28 · rating

Mistral Medium 3 (2505)

1,341.81 rating

Confidence interval: 1,328.38 – 1,355.24

1806 votes

Category
expert
Evaluated configuration
Mistral Medium 3 (2505)
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-28 · rating

Mistral Medium 3 (2505)

1,368.40 rating

Confidence interval: 1,342.10 – 1,394.70

496 votes

Category
french
Evaluated configuration
Mistral Medium 3 (2505)
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-28 · rating

Mistral Medium 3 (2505)

1,377.48 rating

Confidence interval: 1,358.84 – 1,396.11

998 votes

Category
german
Evaluated configuration
Mistral Medium 3 (2505)
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-28 · rating

Mistral Medium 3 (2505)

1,364.38 rating

Confidence interval: 1,358.43 – 1,370.34

13649 votes

Category
hard_prompts
Evaluated configuration
Mistral Medium 3 (2505)
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating

Mistral Medium 3 (2505)

1,373.81 rating

Confidence interval: 1,366.04 – 1,381.57

6364 votes

Category
hard_prompts_english
Evaluated configuration
Mistral Medium 3 (2505)
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating

Mistral Medium 3 (2505)

1,351.60 rating

Confidence interval: 1,342.92 – 1,360.27

5092 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
Mistral Medium 3 (2505)
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-28 · rating

Mistral Medium 3 (2505)

1,335.02 rating

Confidence interval: 1,326.90 – 1,343.14

5598 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
Mistral Medium 3 (2505)
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-28 · rating

Mistral Medium 3 (2505)

1,384.05 rating

Confidence interval: 1,371.25 – 1,396.86

2117 votes

Category
industry_legal_and_government
Evaluated configuration
Mistral Medium 3 (2505)
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-28 · rating

Mistral Medium 3 (2505)

1,380.65 rating

Confidence interval: 1,372.32 – 1,388.98

5632 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
Mistral Medium 3 (2505)
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-28 · rating

Mistral Medium 3 (2505)

1,354.07 rating

Confidence interval: 1,341.89 – 1,366.26

2192 votes

Category
industry_mathematical
Evaluated configuration
Mistral Medium 3 (2505)
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-28 · rating

Mistral Medium 3 (2505)

1,380.97 rating

Confidence interval: 1,367.38 – 1,394.57

1974 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
Mistral Medium 3 (2505)
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-28 · rating

Mistral Medium 3 (2505)

1,385.21 rating

Confidence interval: 1,378.67 – 1,391.74

10552 votes

Category
industry_software_and_it_services
Evaluated configuration
Mistral Medium 3 (2505)
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-28 · rating

Mistral Medium 3 (2505)

1,351.76 rating

Confidence interval: 1,344.25 – 1,359.27

7175 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
Mistral Medium 3 (2505)
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-28 · rating

Mistral Medium 3 (2505)

1,338.38 rating

Confidence interval: 1,331.25 – 1,345.52

7947 votes

Category
instruction_following
Evaluated configuration
Mistral Medium 3 (2505)
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-28 · rating

Mistral Medium 3 (2505)

1,316.04 rating

Confidence interval: 1,296.02 – 1,336.06

916 votes

Category
japanese
Evaluated configuration
Mistral Medium 3 (2505)
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-10-28 · rating

Mistral Medium 3 (2505)

1,303.72 rating

Confidence interval: 1,279.94 – 1,327.49

638 votes

Category
korean
Evaluated configuration
Mistral Medium 3 (2505)
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-28 · rating

Mistral Medium 3 (2505)

1,360.02 rating

Confidence interval: 1,352.04 – 1,368.00

5929 votes

Category
longer_query
Evaluated configuration
Mistral Medium 3 (2505)
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-28 · rating

Mistral Medium 3 (2505)

1,352.54 rating

Confidence interval: 1,340.57 – 1,364.52

2230 votes

Category
math
Evaluated configuration
Mistral Medium 3 (2505)
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-28 · rating

Mistral Medium 3 (2505)

1,383.42 rating

Confidence interval: 1,375.21 – 1,391.63

5713 votes

Category
multi_turn
Evaluated configuration
Mistral Medium 3 (2505)
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-28 · rating

Mistral Medium 3 (2505)

1,353.02 rating

Confidence interval: 1,347.45 – 1,358.58

17661 votes

Category
non_english
Evaluated configuration
Mistral Medium 3 (2505)
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-28 · rating

Mistral Medium 3 (2505)

1,370.11 rating

Confidence interval: 1,365.42 – 1,374.79

32777 votes

Category
overall
Evaluated configuration
Mistral Medium 3 (2505)
Evaluation type
human preference
Arena text preference / polishhuman preference · 2026-10-28 · rating

Mistral Medium 3 (2505)

1,370.36 rating

Confidence interval: 1,358.71 – 1,382.01

2885 votes

Category
polish
Evaluated configuration
Mistral Medium 3 (2505)
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-28 · rating

Mistral Medium 3 (2505)

1,359.23 rating

Confidence interval: 1,348.35 – 1,370.11

2762 votes

Category
russian
Evaluated configuration
Mistral Medium 3 (2505)
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-28 · rating

Mistral Medium 3 (2505)

1,363.93 rating

Confidence interval: 1,338.87 – 1,388.99

565 votes

Category
spanish
Evaluated configuration
Mistral Medium 3 (2505)
Evaluation type
human preference

Run it locally

Inference · memory estimate

Mistral Medium 3 (2505)

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license
Proprietary