deepseek-v3.1-thinking
Not enough verified architecture information for a reliable VRAM estimate.
GeekRadar / The comparison desk
COMPARE.
AI modelsdeepseek-v3.1-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 · 2026-10-15 · rating
deepseek-v3.1-thinking
1,468.56 rating
Confidence interval: 1,450.43 – 1,486.69
1142 votes
- Category
- chinese
- Evaluated configuration
- deepseek-v3.1-thinking
- Evaluation type
- human preference
Arena text preference / codinghuman preference · 2026-10-15 · rating
deepseek-v3.1-thinking
1,416.94 rating
Confidence interval: 1,403.47 – 1,430.40
1874 votes
- Category
- coding
- Evaluated configuration
- deepseek-v3.1-thinking
- Evaluation type
- human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating
deepseek-v3.1-thinking
1,401.02 rating
Confidence interval: 1,386.28 – 1,415.76
1616 votes
- Category
- creative_writing
- Evaluated configuration
- deepseek-v3.1-thinking
- Evaluation type
- human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating
deepseek-v3.1-thinking
1,432.08 rating
Confidence interval: 1,423.11 – 1,441.05
4663 votes
- Category
- english
- Evaluated configuration
- deepseek-v3.1-thinking
- Evaluation type
- human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating
deepseek-v3.1-thinking
1,403.17 rating
Confidence interval: 1,394.20 – 1,412.14
8140 votes
- Category
- exclude_ties
- Evaluated configuration
- deepseek-v3.1-thinking
- Evaluation type
- human preference
Arena text preference / experthuman preference · 2026-10-15 · rating
deepseek-v3.1-thinking
1,404.89 rating
Confidence interval: 1,380.27 – 1,429.50
538 votes
- Category
- expert
- Evaluated configuration
- deepseek-v3.1-thinking
- Evaluation type
- human preference
Arena text preference / frenchhuman preference · 2026-10-15 · rating
deepseek-v3.1-thinking
1,401.44 rating
Confidence interval: 1,357.45 – 1,445.43
183 votes
- Category
- french
- Evaluated configuration
- deepseek-v3.1-thinking
- Evaluation type
- human preference
Arena text preference / germanhuman preference · 2026-10-15 · rating
deepseek-v3.1-thinking
1,407.31 rating
Confidence interval: 1,375.18 – 1,439.45
322 votes
- Category
- german
- Evaluated configuration
- deepseek-v3.1-thinking
- Evaluation type
- human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating
deepseek-v3.1-thinking
1,416.69 rating
Confidence interval: 1,408.07 – 1,425.30
5086 votes
- Category
- hard_prompts
- Evaluated configuration
- deepseek-v3.1-thinking
- Evaluation type
- human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating
deepseek-v3.1-thinking
1,427.13 rating
Confidence interval: 1,414.45 – 1,439.81
2130 votes
- Category
- hard_prompts_english
- Evaluated configuration
- deepseek-v3.1-thinking
- Evaluation type
- human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating
deepseek-v3.1-thinking
1,409.00 rating
Confidence interval: 1,395.61 – 1,422.39
1936 votes
- Category
- industry_business_and_management_and_financial_operations
- Evaluated configuration
- deepseek-v3.1-thinking
- Evaluation type
- human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating
deepseek-v3.1-thinking
1,393.51 rating
Confidence interval: 1,381.02 – 1,406.00
2202 votes
- Category
- industry_entertainment_and_sports_and_media
- Evaluated configuration
- deepseek-v3.1-thinking
- Evaluation type
- human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-15 · rating
deepseek-v3.1-thinking
1,416.50 rating
Confidence interval: 1,395.90 – 1,437.10
771 votes
- Category
- industry_legal_and_government
- Evaluated configuration
- deepseek-v3.1-thinking
- Evaluation type
- human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating
deepseek-v3.1-thinking
1,427.81 rating
Confidence interval: 1,414.17 – 1,441.45
1831 votes
- Category
- industry_life_and_physical_and_social_science
- Evaluated configuration
- deepseek-v3.1-thinking
- Evaluation type
- human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-15 · rating
deepseek-v3.1-thinking
1,437.45 rating
Confidence interval: 1,413.62 – 1,461.28
561 votes
- Category
- industry_mathematical
- Evaluated configuration
- deepseek-v3.1-thinking
- Evaluation type
- human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating
deepseek-v3.1-thinking
1,443.35 rating
Confidence interval: 1,420.82 – 1,465.88
687 votes
- Category
- industry_medicine_and_healthcare
- Evaluated configuration
- deepseek-v3.1-thinking
- Evaluation type
- human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating
deepseek-v3.1-thinking
1,426.97 rating
Confidence interval: 1,416.87 – 1,437.08
3466 votes
- Category
- industry_software_and_it_services
- Evaluated configuration
- deepseek-v3.1-thinking
- Evaluation type
- human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating
deepseek-v3.1-thinking
1,404.50 rating
Confidence interval: 1,393.06 – 1,415.94
2721 votes
- Category
- industry_writing_and_literature_and_language
- Evaluated configuration
- deepseek-v3.1-thinking
- Evaluation type
- human preference
Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating
deepseek-v3.1-thinking
1,400.34 rating
Confidence interval: 1,389.24 – 1,411.44
2832 votes
- Category
- instruction_following
- Evaluated configuration
- deepseek-v3.1-thinking
- Evaluation type
- human preference
Arena text preference / japanesehuman preference · 2026-10-15 · rating
deepseek-v3.1-thinking
1,359.27 rating
Confidence interval: 1,322.73 – 1,395.81
258 votes
- Category
- japanese
- Evaluated configuration
- deepseek-v3.1-thinking
- Evaluation type
- human preference
Arena text preference / koreanhuman preference · 2026-10-15 · rating
deepseek-v3.1-thinking
1,336.41 rating
Confidence interval: 1,304.88 – 1,367.94
373 votes
- Category
- korean
- Evaluated configuration
- deepseek-v3.1-thinking
- Evaluation type
- human preference
Arena text preference / longer_queryhuman preference · 2026-10-15 · rating
deepseek-v3.1-thinking
1,421.73 rating
Confidence interval: 1,409.77 – 1,433.69
2373 votes
- Category
- longer_query
- Evaluated configuration
- deepseek-v3.1-thinking
- Evaluation type
- human preference
Arena text preference / mathhuman preference · 2026-10-15 · rating
deepseek-v3.1-thinking
1,414.86 rating
Confidence interval: 1,392.87 – 1,436.85
658 votes
- Category
- math
- Evaluated configuration
- deepseek-v3.1-thinking
- Evaluation type
- human preference
Arena text preference / multi_turnhuman preference · 2026-10-15 · rating
deepseek-v3.1-thinking
1,407.44 rating
Confidence interval: 1,393.55 – 1,421.33
1815 votes
- Category
- multi_turn
- Evaluated configuration
- deepseek-v3.1-thinking
- Evaluation type
- human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating
deepseek-v3.1-thinking
1,400.01 rating
Confidence interval: 1,392.18 – 1,407.83
6796 votes
- Category
- non_english
- Evaluated configuration
- deepseek-v3.1-thinking
- Evaluation type
- human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating
deepseek-v3.1-thinking
1,416.60 rating
Confidence interval: 1,409.98 – 1,423.22
11459 votes
- Category
- overall
- Evaluated configuration
- deepseek-v3.1-thinking
- Evaluation type
- human preference
Arena text preference / polishhuman preference · 2026-10-15 · rating
deepseek-v3.1-thinking
1,404.87 rating
Confidence interval: 1,384.35 – 1,425.40
787 votes
- Category
- polish
- Evaluated configuration
- deepseek-v3.1-thinking
- Evaluation type
- human preference
Arena text preference / russianhuman preference · 2026-10-15 · rating
deepseek-v3.1-thinking
1,399.29 rating
Confidence interval: 1,379.10 – 1,419.48
727 votes
- Category
- russian
- Evaluated configuration
- deepseek-v3.1-thinking
- Evaluation type
- human preference
Arena text preference / spanishhuman preference · 2026-10-15 · rating
deepseek-v3.1-thinking
1,411.50 rating
Confidence interval: 1,381.55 – 1,441.46
388 votes
- Category
- spanish
- Evaluated configuration
- deepseek-v3.1-thinking
- Evaluation type
- human preference
Run it locally
Capabilities
- Model license
- MIT
0 selected Compare 0