DeepSeek-V4-Pro Preview (High)
Not enough verified architecture information for a reliable VRAM estimate.
GeekRadar / The comparison desk
COMPARE.
AI modelsDeepSeek-V4-Pro Preview (High)
Model evaluation
Capability & preference
Human preference · Arena
Overall text preferences from head-to-head votes. Not factual accuracy or a task-success rate.
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-V4-Pro Preview (High)
1,486.25 rating
Confidence interval: 1,475.37 – 1,497.13
3325 votes
- Category
- chinese
- Evaluated configuration
- DeepSeek-V4-Pro Preview (High)
- Evaluation type
- human preference
Arena text preference / codinghuman preference · 2026-10-28 · rating
DeepSeek-V4-Pro Preview (High)
1,453.14 rating
Confidence interval: 1,447.09 – 1,459.18
16003 votes
- Category
- coding
- Evaluated configuration
- DeepSeek-V4-Pro Preview (High)
- Evaluation type
- human preference
Arena text preference / creative_writinghuman preference · 2026-10-28 · rating
DeepSeek-V4-Pro Preview (High)
1,440.90 rating
Confidence interval: 1,433.62 – 1,448.18
9369 votes
- Category
- creative_writing
- Evaluated configuration
- DeepSeek-V4-Pro Preview (High)
- Evaluation type
- human preference
Arena text preference / englishhuman preference · 2026-10-28 · rating
DeepSeek-V4-Pro Preview (High)
1,454.95 rating
Confidence interval: 1,449.80 – 1,460.10
23986 votes
- Category
- english
- Evaluated configuration
- DeepSeek-V4-Pro Preview (High)
- Evaluation type
- human preference
Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating
DeepSeek-V4-Pro Preview (High)
1,440.69 rating
Confidence interval: 1,435.63 – 1,445.75
41710 votes
- Category
- exclude_ties
- Evaluated configuration
- DeepSeek-V4-Pro Preview (High)
- Evaluation type
- human preference
Arena text preference / experthuman preference · 2026-10-28 · rating
DeepSeek-V4-Pro Preview (High)
1,449.13 rating
Confidence interval: 1,440.58 – 1,457.68
5546 votes
- Category
- expert
- Evaluated configuration
- DeepSeek-V4-Pro Preview (High)
- Evaluation type
- human preference
Arena text preference / frenchhuman preference · 2026-10-28 · rating
DeepSeek-V4-Pro Preview (High)
1,451.15 rating
Confidence interval: 1,436.58 – 1,465.73
2065 votes
- Category
- french
- Evaluated configuration
- DeepSeek-V4-Pro Preview (High)
- Evaluation type
- human preference
Arena text preference / germanhuman preference · 2026-10-28 · rating
DeepSeek-V4-Pro Preview (High)
1,447.89 rating
Confidence interval: 1,428.05 – 1,467.72
922 votes
- Category
- german
- Evaluated configuration
- DeepSeek-V4-Pro Preview (High)
- Evaluation type
- human preference
Arena text preference / hard_promptshuman preference · 2026-10-28 · rating
DeepSeek-V4-Pro Preview (High)
1,452.12 rating
Confidence interval: 1,447.50 – 1,456.74
36907 votes
- Category
- hard_prompts
- Evaluated configuration
- DeepSeek-V4-Pro Preview (High)
- Evaluation type
- human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating
DeepSeek-V4-Pro Preview (High)
1,460.35 rating
Confidence interval: 1,454.43 – 1,466.27
16884 votes
- Category
- hard_prompts_english
- Evaluated configuration
- DeepSeek-V4-Pro Preview (High)
- Evaluation type
- human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating
DeepSeek-V4-Pro Preview (High)
1,427.64 rating
Confidence interval: 1,420.89 – 1,434.38
11029 votes
- Category
- industry_business_and_management_and_financial_operations
- Evaluated configuration
- DeepSeek-V4-Pro Preview (High)
- Evaluation type
- human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-28 · rating
DeepSeek-V4-Pro Preview (High)
1,426.90 rating
Confidence interval: 1,420.29 – 1,433.51
12385 votes
- Category
- industry_entertainment_and_sports_and_media
- Evaluated configuration
- DeepSeek-V4-Pro Preview (High)
- Evaluation type
- human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-28 · rating
DeepSeek-V4-Pro Preview (High)
1,459.37 rating
Confidence interval: 1,449.80 – 1,468.93
4331 votes
- Category
- industry_legal_and_government
- Evaluated configuration
- DeepSeek-V4-Pro Preview (High)
- Evaluation type
- human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-28 · rating
DeepSeek-V4-Pro Preview (High)
1,462.97 rating
Confidence interval: 1,455.85 – 1,470.09
8793 votes
- Category
- industry_life_and_physical_and_social_science
- Evaluated configuration
- DeepSeek-V4-Pro Preview (High)
- Evaluation type
- human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-28 · rating
DeepSeek-V4-Pro Preview (High)
1,453.74 rating
Confidence interval: 1,442.37 – 1,465.11
2993 votes
- Category
- industry_mathematical
- Evaluated configuration
- DeepSeek-V4-Pro Preview (High)
- Evaluation type
- human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-28 · rating
DeepSeek-V4-Pro Preview (High)
1,468.24 rating
Confidence interval: 1,458.13 – 1,478.34
3849 votes
- Category
- industry_medicine_and_healthcare
- Evaluated configuration
- DeepSeek-V4-Pro Preview (High)
- Evaluation type
- human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-28 · rating
DeepSeek-V4-Pro Preview (High)
1,456.67 rating
Confidence interval: 1,451.33 – 1,462.01
22413 votes
- Category
- industry_software_and_it_services
- Evaluated configuration
- DeepSeek-V4-Pro Preview (High)
- Evaluation type
- human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-28 · rating
DeepSeek-V4-Pro Preview (High)
1,437.60 rating
Confidence interval: 1,431.26 – 1,443.95
13705 votes
- Category
- industry_writing_and_literature_and_language
- Evaluated configuration
- DeepSeek-V4-Pro Preview (High)
- Evaluation type
- human preference
Arena text preference / instruction_followinghuman preference · 2026-10-28 · rating
DeepSeek-V4-Pro Preview (High)
1,435.73 rating
Confidence interval: 1,430.09 – 1,441.38
19046 votes
- Category
- instruction_following
- Evaluated configuration
- DeepSeek-V4-Pro Preview (High)
- Evaluation type
- human preference
Arena text preference / japanesehuman preference · 2026-10-28 · rating
DeepSeek-V4-Pro Preview (High)
1,413.23 rating
Confidence interval: 1,389.50 – 1,436.95
691 votes
- Category
- japanese
- Evaluated configuration
- DeepSeek-V4-Pro Preview (High)
- Evaluation type
- human preference
Arena text preference / koreanhuman preference · 2026-10-28 · rating
DeepSeek-V4-Pro Preview (High)
1,424.37 rating
Confidence interval: 1,404.74 – 1,443.99
986 votes
- Category
- korean
- Evaluated configuration
- DeepSeek-V4-Pro Preview (High)
- Evaluation type
- human preference
Arena text preference / longer_queryhuman preference · 2026-10-28 · rating
DeepSeek-V4-Pro Preview (High)
1,445.87 rating
Confidence interval: 1,440.49 – 1,451.24
24713 votes
- Category
- longer_query
- Evaluated configuration
- DeepSeek-V4-Pro Preview (High)
- Evaluation type
- human preference
Arena text preference / mathhuman preference · 2026-10-28 · rating
DeepSeek-V4-Pro Preview (High)
1,453.96 rating
Confidence interval: 1,442.23 – 1,465.70
2714 votes
- Category
- math
- Evaluated configuration
- DeepSeek-V4-Pro Preview (High)
- Evaluation type
- human preference
Arena text preference / multi_turnhuman preference · 2026-10-28 · rating
DeepSeek-V4-Pro Preview (High)
1,445.52 rating
Confidence interval: 1,438.40 – 1,452.64
9688 votes
- Category
- multi_turn
- Evaluated configuration
- DeepSeek-V4-Pro Preview (High)
- Evaluation type
- human preference
Arena text preference / non_englishhuman preference · 2026-10-28 · rating
DeepSeek-V4-Pro Preview (High)
1,431.22 rating
Confidence interval: 1,426.47 – 1,435.98
30880 votes
- Category
- non_english
- Evaluated configuration
- DeepSeek-V4-Pro Preview (High)
- Evaluation type
- human preference
Arena text preference / overallhuman preference · 2026-10-28 · rating
DeepSeek-V4-Pro Preview (High)
1,444.91 rating
Confidence interval: 1,440.97 – 1,448.85
54870 votes
- Category
- overall
- Evaluated configuration
- DeepSeek-V4-Pro Preview (High)
- Evaluation type
- human preference
Arena text preference / polishhuman preference · 2026-10-28 · rating
DeepSeek-V4-Pro Preview (High)
1,454.13 rating
Confidence interval: 1,436.02 – 1,472.24
1090 votes
- Category
- polish
- Evaluated configuration
- DeepSeek-V4-Pro Preview (High)
- Evaluation type
- human preference
Arena text preference / russianhuman preference · 2026-10-28 · rating
DeepSeek-V4-Pro Preview (High)
1,449.27 rating
Confidence interval: 1,441.02 – 1,457.52
6144 votes
- Category
- russian
- Evaluated configuration
- DeepSeek-V4-Pro Preview (High)
- Evaluation type
- human preference
Arena text preference / spanishhuman preference · 2026-10-28 · rating
DeepSeek-V4-Pro Preview (High)
1,448.06 rating
Confidence interval: 1,432.71 – 1,463.42
1680 votes
- Category
- spanish
- Evaluated configuration
- DeepSeek-V4-Pro Preview (High)
- Evaluation type
- human preference
Run it locally
Capabilities
- Model license
- MIT
0 selected Compare 0