GeekRadar / The comparison desk
COMPARE.
AI models
mistral

mistral-medium-2508

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-2508 1424.2 mistral-medium-2508 · 95% interval 1421.7–1426.8 · 95,321 votes · 2026-10-15

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 · 2026-10-15 · rating

mistral-medium-2508

1,447.37 rating

Confidence interval: 1,440.11 – 1,454.63

7170 votes

Category
chinese
Evaluated configuration
mistral-medium-2508
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-15 · rating

mistral-medium-2508

1,433.29 rating

Confidence interval: 1,428.75 – 1,437.82

21571 votes

Category
coding
Evaluated configuration
mistral-medium-2508
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating

mistral-medium-2508

1,390.66 rating

Confidence interval: 1,385.20 – 1,396.12

14034 votes

Category
creative_writing
Evaluated configuration
mistral-medium-2508
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating

mistral-medium-2508

1,437.95 rating

Confidence interval: 1,434.39 – 1,441.51

40859 votes

Category
english
Evaluated configuration
mistral-medium-2508
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating

mistral-medium-2508

1,415.04 rating

Confidence interval: 1,411.45 – 1,418.62

68471 votes

Category
exclude_ties
Evaluated configuration
mistral-medium-2508
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-15 · rating

mistral-medium-2508

1,408.20 rating

Confidence interval: 1,400.71 – 1,415.69

6579 votes

Category
expert
Evaluated configuration
mistral-medium-2508
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-15 · rating

mistral-medium-2508

1,459.15 rating

Confidence interval: 1,445.24 – 1,473.05

2111 votes

Category
french
Evaluated configuration
mistral-medium-2508
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-15 · rating

mistral-medium-2508

1,432.36 rating

Confidence interval: 1,417.73 – 1,446.99

1745 votes

Category
german
Evaluated configuration
mistral-medium-2508
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating

mistral-medium-2508

1,425.62 rating

Confidence interval: 1,422.33 – 1,428.91

52930 votes

Category
hard_prompts
Evaluated configuration
mistral-medium-2508
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating

mistral-medium-2508

1,434.64 rating

Confidence interval: 1,430.21 – 1,439.06

23185 votes

Category
hard_prompts_english
Evaluated configuration
mistral-medium-2508
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating

mistral-medium-2508

1,417.90 rating

Confidence interval: 1,413.01 – 1,422.79

17855 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
mistral-medium-2508
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating

mistral-medium-2508

1,389.62 rating

Confidence interval: 1,384.74 – 1,394.50

18056 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
mistral-medium-2508
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-15 · rating

mistral-medium-2508

1,439.16 rating

Confidence interval: 1,431.72 – 1,446.59

6696 votes

Category
industry_legal_and_government
Evaluated configuration
mistral-medium-2508
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating

mistral-medium-2508

1,444.29 rating

Confidence interval: 1,439.10 – 1,449.47

15294 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
mistral-medium-2508
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-15 · rating

mistral-medium-2508

1,412.16 rating

Confidence interval: 1,403.76 – 1,420.57

5101 votes

Category
industry_mathematical
Evaluated configuration
mistral-medium-2508
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating

mistral-medium-2508

1,449.31 rating

Confidence interval: 1,441.32 – 1,457.30

5996 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
mistral-medium-2508
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating

mistral-medium-2508

1,436.39 rating

Confidence interval: 1,432.59 – 1,440.19

33931 votes

Category
industry_software_and_it_services
Evaluated configuration
mistral-medium-2508
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating

mistral-medium-2508

1,395.25 rating

Confidence interval: 1,390.74 – 1,399.76

21936 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
mistral-medium-2508
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating

mistral-medium-2508

1,397.62 rating

Confidence interval: 1,393.51 – 1,401.73

27636 votes

Category
instruction_following
Evaluated configuration
mistral-medium-2508
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-15 · rating

mistral-medium-2508

1,378.72 rating

Confidence interval: 1,360.74 – 1,396.69

1126 votes

Category
japanese
Evaluated configuration
mistral-medium-2508
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-10-15 · rating

mistral-medium-2508

1,380.58 rating

Confidence interval: 1,366.55 – 1,394.61

1995 votes

Category
korean
Evaluated configuration
mistral-medium-2508
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-15 · rating

mistral-medium-2508

1,405.86 rating

Confidence interval: 1,401.67 – 1,410.04

27810 votes

Category
longer_query
Evaluated configuration
mistral-medium-2508
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-15 · rating

mistral-medium-2508

1,408.49 rating

Confidence interval: 1,400.82 – 1,416.17

5995 votes

Category
math
Evaluated configuration
mistral-medium-2508
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-15 · rating

mistral-medium-2508

1,418.12 rating

Confidence interval: 1,413.03 – 1,423.21

16395 votes

Category
multi_turn
Evaluated configuration
mistral-medium-2508
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating

mistral-medium-2508

1,407.78 rating

Confidence interval: 1,404.57 – 1,410.99

54436 votes

Category
non_english
Evaluated configuration
mistral-medium-2508
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating

mistral-medium-2508

1,424.22 rating

Confidence interval: 1,421.66 – 1,426.78

95321 votes

Category
overall
Evaluated configuration
mistral-medium-2508
Evaluation type
human preference
Arena text preference / polishhuman preference · 2026-10-15 · rating

mistral-medium-2508

1,429.92 rating

Confidence interval: 1,419.42 – 1,440.43

3328 votes

Category
polish
Evaluated configuration
mistral-medium-2508
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-15 · rating

mistral-medium-2508

1,411.30 rating

Confidence interval: 1,405.09 – 1,417.51

9723 votes

Category
russian
Evaluated configuration
mistral-medium-2508
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-15 · rating

mistral-medium-2508

1,433.28 rating

Confidence interval: 1,421.42 – 1,445.13

2868 votes

Category
spanish
Evaluated configuration
mistral-medium-2508
Evaluation type
human preference

Run it locally

Inference · memory estimate

mistral-medium-2508

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license
Proprietary