DeepSeek-V3.2 (Thinking)
Not enough verified architecture information for a reliable VRAM estimate.
GeekRadar / The comparison desk
COMPARE.
AI modelsDeepSeek-V3.2 (Thinking)
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-V3.2 (Thinking)
1,447.90 rating
Confidence interval: 1,437.31 – 1,458.48
3129 votes
- Category
- chinese
- Evaluated configuration
- DeepSeek-V3.2 (Thinking)
- Evaluation type
- human preference
Arena text preference / codinghuman preference · 2026-10-28 · rating
DeepSeek-V3.2 (Thinking)
1,453.18 rating
Confidence interval: 1,446.55 – 1,459.81
8871 votes
- Category
- coding
- Evaluated configuration
- DeepSeek-V3.2 (Thinking)
- Evaluation type
- human preference
Arena text preference / creative_writinghuman preference · 2026-10-28 · rating
DeepSeek-V3.2 (Thinking)
1,388.97 rating
Confidence interval: 1,381.32 – 1,396.62
6409 votes
- Category
- creative_writing
- Evaluated configuration
- DeepSeek-V3.2 (Thinking)
- Evaluation type
- human preference
Arena text preference / englishhuman preference · 2026-10-28 · rating
DeepSeek-V3.2 (Thinking)
1,439.09 rating
Confidence interval: 1,434.01 – 1,444.17
17146 votes
- Category
- english
- Evaluated configuration
- DeepSeek-V3.2 (Thinking)
- Evaluation type
- human preference
Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating
DeepSeek-V3.2 (Thinking)
1,406.89 rating
Confidence interval: 1,401.90 – 1,411.89
29430 votes
- Category
- exclude_ties
- Evaluated configuration
- DeepSeek-V3.2 (Thinking)
- Evaluation type
- human preference
Arena text preference / experthuman preference · 2026-10-28 · rating
DeepSeek-V3.2 (Thinking)
1,431.02 rating
Confidence interval: 1,419.98 – 1,442.06
2880 votes
- Category
- expert
- Evaluated configuration
- DeepSeek-V3.2 (Thinking)
- Evaluation type
- human preference
Arena text preference / frenchhuman preference · 2026-10-28 · rating
DeepSeek-V3.2 (Thinking)
1,422.21 rating
Confidence interval: 1,403.40 – 1,441.01
1045 votes
- Category
- french
- Evaluated configuration
- DeepSeek-V3.2 (Thinking)
- Evaluation type
- human preference
Arena text preference / germanhuman preference · 2026-10-28 · rating
DeepSeek-V3.2 (Thinking)
1,413.47 rating
Confidence interval: 1,392.42 – 1,434.51
748 votes
- Category
- german
- Evaluated configuration
- DeepSeek-V3.2 (Thinking)
- Evaluation type
- human preference
Arena text preference / hard_promptshuman preference · 2026-10-28 · rating
DeepSeek-V3.2 (Thinking)
1,430.48 rating
Confidence interval: 1,425.91 – 1,435.05
23318 votes
- Category
- hard_prompts
- Evaluated configuration
- DeepSeek-V3.2 (Thinking)
- Evaluation type
- human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating
DeepSeek-V3.2 (Thinking)
1,445.61 rating
Confidence interval: 1,439.18 – 1,452.03
9593 votes
- Category
- hard_prompts_english
- Evaluated configuration
- DeepSeek-V3.2 (Thinking)
- Evaluation type
- human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating
DeepSeek-V3.2 (Thinking)
1,409.08 rating
Confidence interval: 1,402.26 – 1,415.90
8101 votes
- Category
- industry_business_and_management_and_financial_operations
- Evaluated configuration
- DeepSeek-V3.2 (Thinking)
- Evaluation type
- human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-28 · rating
DeepSeek-V3.2 (Thinking)
1,386.37 rating
Confidence interval: 1,379.38 – 1,393.36
7745 votes
- Category
- industry_entertainment_and_sports_and_media
- Evaluated configuration
- DeepSeek-V3.2 (Thinking)
- Evaluation type
- human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-28 · rating
DeepSeek-V3.2 (Thinking)
1,418.80 rating
Confidence interval: 1,408.34 – 1,429.25
3085 votes
- Category
- industry_legal_and_government
- Evaluated configuration
- DeepSeek-V3.2 (Thinking)
- Evaluation type
- human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-28 · rating
DeepSeek-V3.2 (Thinking)
1,425.77 rating
Confidence interval: 1,418.44 – 1,433.10
6865 votes
- Category
- industry_life_and_physical_and_social_science
- Evaluated configuration
- DeepSeek-V3.2 (Thinking)
- Evaluation type
- human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-28 · rating
DeepSeek-V3.2 (Thinking)
1,440.02 rating
Confidence interval: 1,426.74 – 1,453.30
1968 votes
- Category
- industry_mathematical
- Evaluated configuration
- DeepSeek-V3.2 (Thinking)
- Evaluation type
- human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-28 · rating
DeepSeek-V3.2 (Thinking)
1,420.28 rating
Confidence interval: 1,409.19 – 1,431.37
2823 votes
- Category
- industry_medicine_and_healthcare
- Evaluated configuration
- DeepSeek-V3.2 (Thinking)
- Evaluation type
- human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-28 · rating
DeepSeek-V3.2 (Thinking)
1,449.37 rating
Confidence interval: 1,443.91 – 1,454.83
14284 votes
- Category
- industry_software_and_it_services
- Evaluated configuration
- DeepSeek-V3.2 (Thinking)
- Evaluation type
- human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-28 · rating
DeepSeek-V3.2 (Thinking)
1,398.27 rating
Confidence interval: 1,391.89 – 1,404.65
9733 votes
- Category
- industry_writing_and_literature_and_language
- Evaluated configuration
- DeepSeek-V3.2 (Thinking)
- Evaluation type
- human preference
Arena text preference / instruction_followinghuman preference · 2026-10-28 · rating
DeepSeek-V3.2 (Thinking)
1,408.35 rating
Confidence interval: 1,402.48 – 1,414.22
11721 votes
- Category
- instruction_following
- Evaluated configuration
- DeepSeek-V3.2 (Thinking)
- Evaluation type
- human preference
Arena text preference / japanesehuman preference · 2026-10-28 · rating
DeepSeek-V3.2 (Thinking)
1,374.87 rating
Confidence interval: 1,345.51 – 1,404.23
385 votes
- Category
- japanese
- Evaluated configuration
- DeepSeek-V3.2 (Thinking)
- Evaluation type
- human preference
Arena text preference / koreanhuman preference · 2026-10-28 · rating
DeepSeek-V3.2 (Thinking)
1,363.36 rating
Confidence interval: 1,340.64 – 1,386.08
700 votes
- Category
- korean
- Evaluated configuration
- DeepSeek-V3.2 (Thinking)
- Evaluation type
- human preference
Arena text preference / longer_queryhuman preference · 2026-10-28 · rating
DeepSeek-V3.2 (Thinking)
1,425.83 rating
Confidence interval: 1,419.89 – 1,431.78
11726 votes
- Category
- longer_query
- Evaluated configuration
- DeepSeek-V3.2 (Thinking)
- Evaluation type
- human preference
Arena text preference / mathhuman preference · 2026-10-28 · rating
DeepSeek-V3.2 (Thinking)
1,422.28 rating
Confidence interval: 1,410.86 – 1,433.71
2631 votes
- Category
- math
- Evaluated configuration
- DeepSeek-V3.2 (Thinking)
- Evaluation type
- human preference
Arena text preference / multi_turnhuman preference · 2026-10-28 · rating
DeepSeek-V3.2 (Thinking)
1,424.08 rating
Confidence interval: 1,416.66 – 1,431.51
6996 votes
- Category
- multi_turn
- Evaluated configuration
- DeepSeek-V3.2 (Thinking)
- Evaluation type
- human preference
Arena text preference / non_englishhuman preference · 2026-10-28 · rating
DeepSeek-V3.2 (Thinking)
1,398.74 rating
Confidence interval: 1,394.32 – 1,403.17
24867 votes
- Category
- non_english
- Evaluated configuration
- DeepSeek-V3.2 (Thinking)
- Evaluation type
- human preference
Arena text preference / overallhuman preference · 2026-10-28 · rating
DeepSeek-V3.2 (Thinking)
1,419.78 rating
Confidence interval: 1,416.23 – 1,423.34
42033 votes
- Category
- overall
- Evaluated configuration
- DeepSeek-V3.2 (Thinking)
- Evaluation type
- human preference
Arena text preference / polishhuman preference · 2026-10-28 · rating
DeepSeek-V3.2 (Thinking)
1,419.22 rating
Confidence interval: 1,402.10 – 1,436.33
1119 votes
- Category
- polish
- Evaluated configuration
- DeepSeek-V3.2 (Thinking)
- Evaluation type
- human preference
Arena text preference / russianhuman preference · 2026-10-28 · rating
DeepSeek-V3.2 (Thinking)
1,407.11 rating
Confidence interval: 1,399.06 – 1,415.16
5416 votes
- Category
- russian
- Evaluated configuration
- DeepSeek-V3.2 (Thinking)
- Evaluation type
- human preference
Arena text preference / spanishhuman preference · 2026-10-28 · rating
DeepSeek-V3.2 (Thinking)
1,415.31 rating
Confidence interval: 1,399.19 – 1,431.43
1365 votes
- Category
- spanish
- Evaluated configuration
- DeepSeek-V3.2 (Thinking)
- Evaluation type
- human preference
Run it locally
Capabilities
- Model license
- MIT
0 selected Compare 0