DeepSeek-V2.5-1210
Stored tensors can include buffers and repeated shared weights. A verified unique parameter count is unavailable. Checkpoint size is not the memory required to run the model.
GeekRadar / The comparison desk
COMPARE.
AI modelsDeepSeek-V2.5-1210
AI model
- Context window
- 163,840 tokens
- Open weights
- Yes
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-V2.5-1210
1,317.54 rating
Confidence interval: 1,291.51 – 1,343.56
446 votes
- Category
- chinese
- Evaluated configuration
- deepseek-v2.5-1210
- Evaluation type
- human preference
Arena text preference / codinghuman preference · 2026-10-15 · rating
DeepSeek-V2.5-1210
1,309.23 rating
Confidence interval: 1,292.50 – 1,325.96
1079 votes
- Category
- coding
- Evaluated configuration
- deepseek-v2.5-1210
- Evaluation type
- human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating
DeepSeek-V2.5-1210
1,284.95 rating
Confidence interval: 1,267.76 – 1,302.14
1090 votes
- Category
- creative_writing
- Evaluated configuration
- deepseek-v2.5-1210
- Evaluation type
- human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating
DeepSeek-V2.5-1210
1,309.51 rating
Confidence interval: 1,299.21 – 1,319.81
3776 votes
- Category
- english
- Evaluated configuration
- deepseek-v2.5-1210
- Evaluation type
- human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating
DeepSeek-V2.5-1210
1,228.66 rating
Confidence interval: 1,215.64 – 1,241.68
4213 votes
- Category
- exclude_ties
- Evaluated configuration
- deepseek-v2.5-1210
- Evaluation type
- human preference
Arena text preference / experthuman preference · 2026-10-15 · rating
DeepSeek-V2.5-1210
1,266.03 rating
Confidence interval: 1,240.08 – 1,291.98
441 votes
- Category
- expert
- Evaluated configuration
- deepseek-v2.5-1210
- Evaluation type
- human preference
Arena text preference / germanhuman preference · 2026-10-15 · rating
DeepSeek-V2.5-1210
1,258.13 rating
Confidence interval: 1,212.47 – 1,303.79
140 votes
- Category
- german
- Evaluated configuration
- deepseek-v2.5-1210
- Evaluation type
- human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating
DeepSeek-V2.5-1210
1,289.07 rating
Confidence interval: 1,275.86 – 1,302.28
1790 votes
- Category
- hard_prompts
- Evaluated configuration
- deepseek-v2.5-1210
- Evaluation type
- human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating
DeepSeek-V2.5-1210
1,298.75 rating
Confidence interval: 1,282.48 – 1,315.01
1132 votes
- Category
- hard_prompts_english
- Evaluated configuration
- deepseek-v2.5-1210
- Evaluation type
- human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating
DeepSeek-V2.5-1210
1,259.22 rating
Confidence interval: 1,237.89 – 1,280.56
715 votes
- Category
- industry_business_and_management_and_financial_operations
- Evaluated configuration
- deepseek-v2.5-1210
- Evaluation type
- human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating
DeepSeek-V2.5-1210
1,275.08 rating
Confidence interval: 1,259.14 – 1,291.02
1216 votes
- Category
- industry_entertainment_and_sports_and_media
- Evaluated configuration
- deepseek-v2.5-1210
- Evaluation type
- human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-15 · rating
DeepSeek-V2.5-1210
1,278.41 rating
Confidence interval: 1,249.21 – 1,307.62
375 votes
- Category
- industry_legal_and_government
- Evaluated configuration
- deepseek-v2.5-1210
- Evaluation type
- human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating
DeepSeek-V2.5-1210
1,288.16 rating
Confidence interval: 1,271.47 – 1,304.85
1137 votes
- Category
- industry_life_and_physical_and_social_science
- Evaluated configuration
- deepseek-v2.5-1210
- Evaluation type
- human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-15 · rating
DeepSeek-V2.5-1210
1,278.15 rating
Confidence interval: 1,260.45 – 1,295.84
893 votes
- Category
- industry_mathematical
- Evaluated configuration
- deepseek-v2.5-1210
- Evaluation type
- human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating
DeepSeek-V2.5-1210
1,291.95 rating
Confidence interval: 1,259.60 – 1,324.30
291 votes
- Category
- industry_medicine_and_healthcare
- Evaluated configuration
- deepseek-v2.5-1210
- Evaluation type
- human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating
DeepSeek-V2.5-1210
1,313.17 rating
Confidence interval: 1,299.17 – 1,327.16
1688 votes
- Category
- industry_software_and_it_services
- Evaluated configuration
- deepseek-v2.5-1210
- Evaluation type
- human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating
DeepSeek-V2.5-1210
1,283.06 rating
Confidence interval: 1,269.74 – 1,296.37
1930 votes
- Category
- industry_writing_and_literature_and_language
- Evaluated configuration
- deepseek-v2.5-1210
- Evaluation type
- human preference
Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating
DeepSeek-V2.5-1210
1,279.49 rating
Confidence interval: 1,268.34 – 1,290.63
2970 votes
- Category
- instruction_following
- Evaluated configuration
- deepseek-v2.5-1210
- Evaluation type
- human preference
Arena text preference / japanesehuman preference · 2026-10-15 · rating
DeepSeek-V2.5-1210
1,227.80 rating
Confidence interval: 1,186.46 – 1,269.14
161 votes
- Category
- japanese
- Evaluated configuration
- deepseek-v2.5-1210
- Evaluation type
- human preference
Arena text preference / longer_queryhuman preference · 2026-10-15 · rating
DeepSeek-V2.5-1210
1,300.52 rating
Confidence interval: 1,283.40 – 1,317.65
1037 votes
- Category
- longer_query
- Evaluated configuration
- deepseek-v2.5-1210
- Evaluation type
- human preference
Arena text preference / mathhuman preference · 2026-10-15 · rating
DeepSeek-V2.5-1210
1,288.42 rating
Confidence interval: 1,271.81 – 1,305.03
1031 votes
- Category
- math
- Evaluated configuration
- deepseek-v2.5-1210
- Evaluation type
- human preference
Arena text preference / multi_turnhuman preference · 2026-10-15 · rating
DeepSeek-V2.5-1210
1,296.62 rating
Confidence interval: 1,280.34 – 1,312.90
1178 votes
- Category
- multi_turn
- Evaluated configuration
- deepseek-v2.5-1210
- Evaluation type
- human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating
DeepSeek-V2.5-1210
1,273.26 rating
Confidence interval: 1,262.19 – 1,284.34
3019 votes
- Category
- non_english
- Evaluated configuration
- deepseek-v2.5-1210
- Evaluation type
- human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating
DeepSeek-V2.5-1210
1,293.89 rating
Confidence interval: 1,285.67 – 1,302.11
6795 votes
- Category
- overall
- Evaluated configuration
- deepseek-v2.5-1210
- Evaluation type
- human preference
Arena text preference / russianhuman preference · 2026-10-15 · rating
DeepSeek-V2.5-1210
1,289.26 rating
Confidence interval: 1,271.98 – 1,306.55
1034 votes
- Category
- russian
- Evaluated configuration
- deepseek-v2.5-1210
- Evaluation type
- human preference
Run it locally
Capabilities
- Context window
- 163,840 tokens
- Open weights
- Yes
- Model license
- other
- Checkpoint tensor payload
- 439.10 GiB
Input modalities
- text
Output modalities
- text
Processing & platform
- Architecture
- deepseek_v2
0 selected Compare 0