GeekRadar / The comparison desk
COMPARE.
AI models
mistral

Magistral Medium (2506)

Model evaluation

Capability & preference

Human preference · Arena

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

Magistral Medium (2506) 1254.8 Magistral Medium (2506) · 95% interval 1248.4–1261.3 · 11,513 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

Magistral Medium (2506)

1,227.13 rating

Confidence interval: 1,203.34 – 1,250.92

726 votes

Category
chinese
Evaluated configuration
Magistral Medium (2506)
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-28 · rating

Magistral Medium (2506)

1,318.77 rating

Confidence interval: 1,306.35 – 1,331.19

2245 votes

Category
coding
Evaluated configuration
Magistral Medium (2506)
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-28 · rating

Magistral Medium (2506)

1,244.80 rating

Confidence interval: 1,228.39 – 1,261.20

1350 votes

Category
creative_writing
Evaluated configuration
Magistral Medium (2506)
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-28 · rating

Magistral Medium (2506)

1,273.91 rating

Confidence interval: 1,264.53 – 1,283.30

4462 votes

Category
english
Evaluated configuration
Magistral Medium (2506)
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating

Magistral Medium (2506)

1,177.52 rating

Confidence interval: 1,168.56 – 1,186.48

8654 votes

Category
exclude_ties
Evaluated configuration
Magistral Medium (2506)
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-28 · rating

Magistral Medium (2506)

1,223.26 rating

Confidence interval: 1,196.70 – 1,249.81

565 votes

Category
expert
Evaluated configuration
Magistral Medium (2506)
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-28 · rating

Magistral Medium (2506)

1,268.89 rating

Confidence interval: 1,222.53 – 1,315.26

183 votes

Category
french
Evaluated configuration
Magistral Medium (2506)
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-28 · rating

Magistral Medium (2506)

1,247.61 rating

Confidence interval: 1,214.93 – 1,280.29

361 votes

Category
german
Evaluated configuration
Magistral Medium (2506)
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-28 · rating

Magistral Medium (2506)

1,266.78 rating

Confidence interval: 1,258.42 – 1,275.14

5640 votes

Category
hard_prompts
Evaluated configuration
Magistral Medium (2506)
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating

Magistral Medium (2506)

1,284.19 rating

Confidence interval: 1,271.67 – 1,296.71

2396 votes

Category
hard_prompts_english
Evaluated configuration
Magistral Medium (2506)
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating

Magistral Medium (2506)

1,222.97 rating

Confidence interval: 1,209.49 – 1,236.45

2217 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
Magistral Medium (2506)
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-28 · rating

Magistral Medium (2506)

1,241.45 rating

Confidence interval: 1,227.98 – 1,254.91

1974 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
Magistral Medium (2506)
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-28 · rating

Magistral Medium (2506)

1,231.67 rating

Confidence interval: 1,209.34 – 1,253.99

842 votes

Category
industry_legal_and_government
Evaluated configuration
Magistral Medium (2506)
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-28 · rating

Magistral Medium (2506)

1,233.52 rating

Confidence interval: 1,218.70 – 1,248.33

1929 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
Magistral Medium (2506)
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-28 · rating

Magistral Medium (2506)

1,255.40 rating

Confidence interval: 1,229.00 – 1,281.80

572 votes

Category
industry_mathematical
Evaluated configuration
Magistral Medium (2506)
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-28 · rating

Magistral Medium (2506)

1,245.55 rating

Confidence interval: 1,222.17 – 1,268.93

775 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
Magistral Medium (2506)
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-28 · rating

Magistral Medium (2506)

1,283.84 rating

Confidence interval: 1,274.11 – 1,293.57

4008 votes

Category
industry_software_and_it_services
Evaluated configuration
Magistral Medium (2506)
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-28 · rating

Magistral Medium (2506)

1,256.85 rating

Confidence interval: 1,244.49 – 1,269.21

2393 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
Magistral Medium (2506)
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-28 · rating

Magistral Medium (2506)

1,253.54 rating

Confidence interval: 1,242.63 – 1,264.44

3115 votes

Category
instruction_following
Evaluated configuration
Magistral Medium (2506)
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-28 · rating

Magistral Medium (2506)

1,175.60 rating

Confidence interval: 1,140.93 – 1,210.27

349 votes

Category
japanese
Evaluated configuration
Magistral Medium (2506)
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-10-28 · rating

Magistral Medium (2506)

1,124.83 rating

Confidence interval: 1,076.91 – 1,172.75

194 votes

Category
korean
Evaluated configuration
Magistral Medium (2506)
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-28 · rating

Magistral Medium (2506)

1,294.14 rating

Confidence interval: 1,281.89 – 1,306.40

2402 votes

Category
longer_query
Evaluated configuration
Magistral Medium (2506)
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-28 · rating

Magistral Medium (2506)

1,249.92 rating

Confidence interval: 1,223.99 – 1,275.85

552 votes

Category
math
Evaluated configuration
Magistral Medium (2506)
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-28 · rating

Magistral Medium (2506)

1,274.84 rating

Confidence interval: 1,259.57 – 1,290.12

1599 votes

Category
multi_turn
Evaluated configuration
Magistral Medium (2506)
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-28 · rating

Magistral Medium (2506)

1,232.07 rating

Confidence interval: 1,224.29 – 1,239.85

7051 votes

Category
non_english
Evaluated configuration
Magistral Medium (2506)
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-28 · rating

Magistral Medium (2506)

1,254.83 rating

Confidence interval: 1,248.39 – 1,261.27

11513 votes

Category
overall
Evaluated configuration
Magistral Medium (2506)
Evaluation type
human preference
Arena text preference / polishhuman preference · 2026-10-28 · rating

Magistral Medium (2506)

1,248.14 rating

Confidence interval: 1,232.00 – 1,264.27

1618 votes

Category
polish
Evaluated configuration
Magistral Medium (2506)
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-28 · rating

Magistral Medium (2506)

1,223.52 rating

Confidence interval: 1,204.18 – 1,242.86

922 votes

Category
russian
Evaluated configuration
Magistral Medium (2506)
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-28 · rating

Magistral Medium (2506)

1,270.56 rating

Confidence interval: 1,231.45 – 1,309.68

269 votes

Category
spanish
Evaluated configuration
Magistral Medium (2506)
Evaluation type
human preference

Run it locally

Inference · memory estimate

Magistral Medium (2506)

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license
Proprietary