GeekRadar / The comparison desk
COMPARE.
AI models
deepseek

DeepSeek-V3 (0324)

Model evaluation

Capability & preference

Human preference · Arena

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

DeepSeek-V3 (0324) 1374.6 DeepSeek-V3 (0324) · 95% interval 1370.8–1378.5 · 44,977 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

DeepSeek-V3 (0324)

1,390.12 rating

Confidence interval: 1,379.25 – 1,400.99

3036 votes

Category
chinese
Evaluated configuration
DeepSeek-V3 (0324)
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-28 · rating

DeepSeek-V3 (0324)

1,367.75 rating

Confidence interval: 1,360.99 – 1,374.51

8333 votes

Category
coding
Evaluated configuration
DeepSeek-V3 (0324)
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-28 · rating

DeepSeek-V3 (0324)

1,363.97 rating

Confidence interval: 1,356.00 – 1,371.95

5992 votes

Category
creative_writing
Evaluated configuration
DeepSeek-V3 (0324)
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-28 · rating

DeepSeek-V3 (0324)

1,388.93 rating

Confidence interval: 1,383.98 – 1,393.89

21421 votes

Category
english
Evaluated configuration
DeepSeek-V3 (0324)
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating

DeepSeek-V3 (0324)

1,345.28 rating

Confidence interval: 1,339.94 – 1,350.62

31456 votes

Category
exclude_ties
Evaluated configuration
DeepSeek-V3 (0324)
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-28 · rating

DeepSeek-V3 (0324)

1,350.43 rating

Confidence interval: 1,338.14 – 1,362.72

2351 votes

Category
expert
Evaluated configuration
DeepSeek-V3 (0324)
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-28 · rating

DeepSeek-V3 (0324)

1,385.50 rating

Confidence interval: 1,360.99 – 1,410.01

594 votes

Category
french
Evaluated configuration
DeepSeek-V3 (0324)
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-28 · rating

DeepSeek-V3 (0324)

1,373.78 rating

Confidence interval: 1,356.34 – 1,391.22

1224 votes

Category
german
Evaluated configuration
DeepSeek-V3 (0324)
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-28 · rating

DeepSeek-V3 (0324)

1,364.89 rating

Confidence interval: 1,359.71 – 1,370.08

18590 votes

Category
hard_prompts
Evaluated configuration
DeepSeek-V3 (0324)
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating

DeepSeek-V3 (0324)

1,378.93 rating

Confidence interval: 1,372.28 – 1,385.59

9014 votes

Category
hard_prompts_english
Evaluated configuration
DeepSeek-V3 (0324)
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating

DeepSeek-V3 (0324)

1,349.11 rating

Confidence interval: 1,341.61 – 1,356.61

6857 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
DeepSeek-V3 (0324)
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-28 · rating

DeepSeek-V3 (0324)

1,357.05 rating

Confidence interval: 1,350.13 – 1,363.96

8019 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
DeepSeek-V3 (0324)
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-28 · rating

DeepSeek-V3 (0324)

1,379.11 rating

Confidence interval: 1,367.97 – 1,390.25

2907 votes

Category
industry_legal_and_government
Evaluated configuration
DeepSeek-V3 (0324)
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-28 · rating

DeepSeek-V3 (0324)

1,381.55 rating

Confidence interval: 1,374.41 – 1,388.69

7630 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
DeepSeek-V3 (0324)
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-28 · rating

DeepSeek-V3 (0324)

1,364.10 rating

Confidence interval: 1,353.42 – 1,374.79

2947 votes

Category
industry_mathematical
Evaluated configuration
DeepSeek-V3 (0324)
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-28 · rating

DeepSeek-V3 (0324)

1,383.00 rating

Confidence interval: 1,371.37 – 1,394.63

2658 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
DeepSeek-V3 (0324)
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-28 · rating

DeepSeek-V3 (0324)

1,377.90 rating

Confidence interval: 1,372.26 – 1,383.53

14363 votes

Category
industry_software_and_it_services
Evaluated configuration
DeepSeek-V3 (0324)
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-28 · rating

DeepSeek-V3 (0324)

1,364.75 rating

Confidence interval: 1,358.38 – 1,371.12

10288 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
DeepSeek-V3 (0324)
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-28 · rating

DeepSeek-V3 (0324)

1,344.49 rating

Confidence interval: 1,338.64 – 1,350.35

12412 votes

Category
instruction_following
Evaluated configuration
DeepSeek-V3 (0324)
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-28 · rating

DeepSeek-V3 (0324)

1,333.08 rating

Confidence interval: 1,314.17 – 1,352.00

1056 votes

Category
japanese
Evaluated configuration
DeepSeek-V3 (0324)
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-10-28 · rating

DeepSeek-V3 (0324)

1,319.14 rating

Confidence interval: 1,299.61 – 1,338.66

970 votes

Category
korean
Evaluated configuration
DeepSeek-V3 (0324)
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-28 · rating

DeepSeek-V3 (0324)

1,352.03 rating

Confidence interval: 1,345.34 – 1,358.73

8535 votes

Category
longer_query
Evaluated configuration
DeepSeek-V3 (0324)
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-28 · rating

DeepSeek-V3 (0324)

1,372.52 rating

Confidence interval: 1,362.23 – 1,382.82

3167 votes

Category
math
Evaluated configuration
DeepSeek-V3 (0324)
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-28 · rating

DeepSeek-V3 (0324)

1,388.59 rating

Confidence interval: 1,381.40 – 1,395.79

7685 votes

Category
multi_turn
Evaluated configuration
DeepSeek-V3 (0324)
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-28 · rating

DeepSeek-V3 (0324)

1,357.84 rating

Confidence interval: 1,353.06 – 1,362.62

23551 votes

Category
non_english
Evaluated configuration
DeepSeek-V3 (0324)
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-28 · rating

DeepSeek-V3 (0324)

1,374.61 rating

Confidence interval: 1,370.76 – 1,378.47

44977 votes

Category
overall
Evaluated configuration
DeepSeek-V3 (0324)
Evaluation type
human preference
Arena text preference / polishhuman preference · 2026-10-28 · rating

DeepSeek-V3 (0324)

1,368.66 rating

Confidence interval: 1,356.69 – 1,380.63

2735 votes

Category
polish
Evaluated configuration
DeepSeek-V3 (0324)
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-28 · rating

DeepSeek-V3 (0324)

1,373.10 rating

Confidence interval: 1,363.57 – 1,382.63

3573 votes

Category
russian
Evaluated configuration
DeepSeek-V3 (0324)
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-28 · rating

DeepSeek-V3 (0324)

1,358.01 rating

Confidence interval: 1,338.72 – 1,377.30

987 votes

Category
spanish
Evaluated configuration
DeepSeek-V3 (0324)
Evaluation type
human preference

Run it locally

Inference · memory estimate

DeepSeek-V3 (0324)

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license
MIT