GeekRadar / The comparison desk
COMPARE.
AI models
mistral

Mistral Medium

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 1165.1 Mistral Medium · 95% interval 1159.7–1170.5 · 34,550 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

1,109.02 rating

Confidence interval: 1,095.47 – 1,122.57

3014 votes

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

Mistral Medium

1,161.77 rating

Confidence interval: 1,151.43 – 1,172.11

5149 votes

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

Mistral Medium

1,157.40 rating

Confidence interval: 1,146.64 – 1,168.15

5549 votes

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

Mistral Medium

1,190.27 rating

Confidence interval: 1,183.86 – 1,196.68

23846 votes

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

Mistral Medium

1,030.40 rating

Confidence interval: 1,022.45 – 1,038.34

23596 votes

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

Mistral Medium

1,125.08 rating

Confidence interval: 1,108.01 – 1,142.15

1357 votes

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

Mistral Medium

1,180.08 rating

Confidence interval: 1,152.82 – 1,207.35

721 votes

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

Mistral Medium

1,154.69 rating

Confidence interval: 1,132.19 – 1,177.19

852 votes

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

Mistral Medium

1,149.13 rating

Confidence interval: 1,140.54 – 1,157.73

8650 votes

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

Mistral Medium

1,157.76 rating

Confidence interval: 1,148.03 – 1,167.48

6398 votes

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

Mistral Medium

1,119.46 rating

Confidence interval: 1,106.86 – 1,132.06

3224 votes

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

Mistral Medium

1,137.71 rating

Confidence interval: 1,127.12 – 1,148.29

5402 votes

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

Mistral Medium

1,169.96 rating

Confidence interval: 1,153.59 – 1,186.33

1811 votes

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

Mistral Medium

1,158.28 rating

Confidence interval: 1,147.81 – 1,168.74

6062 votes

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

Mistral Medium

1,177.87 rating

Confidence interval: 1,166.54 – 1,189.20

3806 votes

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

Mistral Medium

1,132.11 rating

Confidence interval: 1,116.02 – 1,148.20

1925 votes

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

Mistral Medium

1,153.60 rating

Confidence interval: 1,144.75 – 1,162.45

9016 votes

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

Mistral Medium

1,171.97 rating

Confidence interval: 1,162.67 – 1,181.27

8402 votes

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

Mistral Medium

1,151.63 rating

Confidence interval: 1,143.56 – 1,159.70

11466 votes

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

Mistral Medium

1,012.63 rating

Confidence interval: 978.83 – 1,046.42

337 votes

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

Mistral Medium

1,032.05 rating

Confidence interval: 996.32 – 1,067.78

268 votes

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

Mistral Medium

1,153.16 rating

Confidence interval: 1,140.41 – 1,165.91

2718 votes

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

Mistral Medium

1,179.80 rating

Confidence interval: 1,169.07 – 1,190.53

4406 votes

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

Mistral Medium

1,147.25 rating

Confidence interval: 1,136.01 – 1,158.49

4621 votes

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

Mistral Medium

1,128.53 rating

Confidence interval: 1,120.28 – 1,136.78

10704 votes

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

Mistral Medium

1,165.11 rating

Confidence interval: 1,159.69 – 1,170.53

34550 votes

Category
overall
Evaluated configuration
Mistral Medium
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-28 · rating

Mistral Medium

1,167.47 rating

Confidence interval: 1,152.52 – 1,182.43

1757 votes

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

Mistral Medium

1,143.38 rating

Confidence interval: 1,111.32 – 1,175.44

439 votes

Category
spanish
Evaluated configuration
Mistral Medium
Evaluation type
human preference

Run it locally

Inference · memory estimate

Mistral Medium

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license
Proprietary