kimi-k2.5-thinking
Not enough verified architecture information for a reliable VRAM estimate.
GeekRadar / The comparison desk
COMPARE.
AI modelskimi-k2.5-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
kimi-k2.5-thinking
1,494.52 rating
Confidence interval: 1,485.43 – 1,503.61
4615 votes
- Category
- chinese
- Evaluated configuration
- kimi-k2.5-thinking
- Evaluation type
- human preference
Arena text preference / codinghuman preference · 2026-10-15 · rating
kimi-k2.5-thinking
1,473.84 rating
Confidence interval: 1,468.71 – 1,478.98
20765 votes
- Category
- coding
- Evaluated configuration
- kimi-k2.5-thinking
- Evaluation type
- human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating
kimi-k2.5-thinking
1,422.24 rating
Confidence interval: 1,416.09 – 1,428.40
12275 votes
- Category
- creative_writing
- Evaluated configuration
- kimi-k2.5-thinking
- Evaluation type
- human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating
kimi-k2.5-thinking
1,453.42 rating
Confidence interval: 1,449.02 – 1,457.82
31761 votes
- Category
- english
- Evaluated configuration
- kimi-k2.5-thinking
- Evaluation type
- human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating
kimi-k2.5-thinking
1,442.13 rating
Confidence interval: 1,437.73 – 1,446.53
54445 votes
- Category
- exclude_ties
- Evaluated configuration
- kimi-k2.5-thinking
- Evaluation type
- human preference
Arena text preference / experthuman preference · 2026-10-15 · rating
kimi-k2.5-thinking
1,465.48 rating
Confidence interval: 1,457.91 – 1,473.05
7018 votes
- Category
- expert
- Evaluated configuration
- kimi-k2.5-thinking
- Evaluation type
- human preference
Arena text preference / frenchhuman preference · 2026-10-15 · rating
kimi-k2.5-thinking
1,453.54 rating
Confidence interval: 1,440.51 – 1,466.57
2630 votes
- Category
- french
- Evaluated configuration
- kimi-k2.5-thinking
- Evaluation type
- human preference
Arena text preference / germanhuman preference · 2026-10-15 · rating
kimi-k2.5-thinking
1,441.12 rating
Confidence interval: 1,423.68 – 1,458.56
1284 votes
- Category
- german
- Evaluated configuration
- kimi-k2.5-thinking
- Evaluation type
- human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating
kimi-k2.5-thinking
1,453.30 rating
Confidence interval: 1,449.36 – 1,457.24
47926 votes
- Category
- hard_prompts
- Evaluated configuration
- kimi-k2.5-thinking
- Evaluation type
- human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating
kimi-k2.5-thinking
1,460.22 rating
Confidence interval: 1,455.05 – 1,465.38
21053 votes
- Category
- hard_prompts_english
- Evaluated configuration
- kimi-k2.5-thinking
- Evaluation type
- human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating
kimi-k2.5-thinking
1,427.03 rating
Confidence interval: 1,421.32 – 1,432.74
14416 votes
- Category
- industry_business_and_management_and_financial_operations
- Evaluated configuration
- kimi-k2.5-thinking
- Evaluation type
- human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating
kimi-k2.5-thinking
1,417.08 rating
Confidence interval: 1,411.42 – 1,422.73
15507 votes
- Category
- industry_entertainment_and_sports_and_media
- Evaluated configuration
- kimi-k2.5-thinking
- Evaluation type
- human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-15 · rating
kimi-k2.5-thinking
1,440.79 rating
Confidence interval: 1,432.53 – 1,449.05
5653 votes
- Category
- industry_legal_and_government
- Evaluated configuration
- kimi-k2.5-thinking
- Evaluation type
- human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating
kimi-k2.5-thinking
1,458.45 rating
Confidence interval: 1,452.42 – 1,464.48
11814 votes
- Category
- industry_life_and_physical_and_social_science
- Evaluated configuration
- kimi-k2.5-thinking
- Evaluation type
- human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-15 · rating
kimi-k2.5-thinking
1,473.27 rating
Confidence interval: 1,463.64 – 1,482.90
4063 votes
- Category
- industry_mathematical
- Evaluated configuration
- kimi-k2.5-thinking
- Evaluation type
- human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating
kimi-k2.5-thinking
1,447.41 rating
Confidence interval: 1,438.68 – 1,456.13
5172 votes
- Category
- industry_medicine_and_healthcare
- Evaluated configuration
- kimi-k2.5-thinking
- Evaluation type
- human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating
kimi-k2.5-thinking
1,469.11 rating
Confidence interval: 1,464.59 – 1,473.63
29773 votes
- Category
- industry_software_and_it_services
- Evaluated configuration
- kimi-k2.5-thinking
- Evaluation type
- human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating
kimi-k2.5-thinking
1,425.24 rating
Confidence interval: 1,419.90 – 1,430.57
17937 votes
- Category
- industry_writing_and_literature_and_language
- Evaluated configuration
- kimi-k2.5-thinking
- Evaluation type
- human preference
Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating
kimi-k2.5-thinking
1,430.66 rating
Confidence interval: 1,425.84 – 1,435.47
24213 votes
- Category
- instruction_following
- Evaluated configuration
- kimi-k2.5-thinking
- Evaluation type
- human preference
Arena text preference / japanesehuman preference · 2026-10-15 · rating
kimi-k2.5-thinking
1,422.31 rating
Confidence interval: 1,400.70 – 1,443.92
851 votes
- Category
- japanese
- Evaluated configuration
- kimi-k2.5-thinking
- Evaluation type
- human preference
Arena text preference / koreanhuman preference · 2026-10-15 · rating
kimi-k2.5-thinking
1,411.15 rating
Confidence interval: 1,393.98 – 1,428.32
1359 votes
- Category
- korean
- Evaluated configuration
- kimi-k2.5-thinking
- Evaluation type
- human preference
Arena text preference / longer_queryhuman preference · 2026-10-15 · rating
kimi-k2.5-thinking
1,444.48 rating
Confidence interval: 1,439.77 – 1,449.18
29837 votes
- Category
- longer_query
- Evaluated configuration
- kimi-k2.5-thinking
- Evaluation type
- human preference
Arena text preference / mathhuman preference · 2026-10-15 · rating
kimi-k2.5-thinking
1,469.83 rating
Confidence interval: 1,460.51 – 1,479.16
4172 votes
- Category
- math
- Evaluated configuration
- kimi-k2.5-thinking
- Evaluation type
- human preference
Arena text preference / multi_turnhuman preference · 2026-10-15 · rating
kimi-k2.5-thinking
1,444.03 rating
Confidence interval: 1,438.07 – 1,449.99
13161 votes
- Category
- multi_turn
- Evaluated configuration
- kimi-k2.5-thinking
- Evaluation type
- human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating
kimi-k2.5-thinking
1,432.64 rating
Confidence interval: 1,428.71 – 1,436.57
42995 votes
- Category
- non_english
- Evaluated configuration
- kimi-k2.5-thinking
- Evaluation type
- human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating
kimi-k2.5-thinking
1,445.02 rating
Confidence interval: 1,441.76 – 1,448.27
74759 votes
- Category
- overall
- Evaluated configuration
- kimi-k2.5-thinking
- Evaluation type
- human preference
Arena text preference / polishhuman preference · 2026-10-15 · rating
kimi-k2.5-thinking
1,445.46 rating
Confidence interval: 1,430.62 – 1,460.30
1522 votes
- Category
- polish
- Evaluated configuration
- kimi-k2.5-thinking
- Evaluation type
- human preference
Arena text preference / russianhuman preference · 2026-10-15 · rating
kimi-k2.5-thinking
1,435.31 rating
Confidence interval: 1,428.57 – 1,442.06
8839 votes
- Category
- russian
- Evaluated configuration
- kimi-k2.5-thinking
- Evaluation type
- human preference
Arena text preference / spanishhuman preference · 2026-10-15 · rating
kimi-k2.5-thinking
1,449.21 rating
Confidence interval: 1,436.47 – 1,461.95
2525 votes
- Category
- spanish
- Evaluated configuration
- kimi-k2.5-thinking
- Evaluation type
- human preference
Run it locally
Capabilities
- Model license
- Modified MIT
0 selected Compare 0