GeekRadar / The comparison desk
COMPARE.
AI models
anthropic

Claude Sonnet 4.5 (20250929)

Model evaluation

Capability & preference

Human preference · Arena

Overall text preferences from head-to-head votes. Not factual accuracy or a task-success rate.

Claude Sonnet 4.5 (20250929) 1438.5 Claude Sonnet 4.5 (20250929) · 95% interval 1435.7–1441.4 · 82,621 votes · 2026-10-28

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

Claude Sonnet 4.5 (20250929)

1,459.43 rating

Confidence interval: 1,451.26 – 1,467.60

5884 votes

Category
chinese
Evaluated configuration
Claude Sonnet 4.5 (20250929)
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-28 · rating

Claude Sonnet 4.5 (20250929)

1,484.71 rating

Confidence interval: 1,479.73 – 1,489.69

19749 votes

Category
coding
Evaluated configuration
Claude Sonnet 4.5 (20250929)
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-28 · rating

Claude Sonnet 4.5 (20250929)

1,441.78 rating

Confidence interval: 1,435.80 – 1,447.77

12398 votes

Category
creative_writing
Evaluated configuration
Claude Sonnet 4.5 (20250929)
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-28 · rating

Claude Sonnet 4.5 (20250929)

1,450.06 rating

Confidence interval: 1,446.12 – 1,454.00

35717 votes

Category
english
Evaluated configuration
Claude Sonnet 4.5 (20250929)
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating

Claude Sonnet 4.5 (20250929)

1,433.95 rating

Confidence interval: 1,430.11 – 1,437.79

60358 votes

Category
exclude_ties
Evaluated configuration
Claude Sonnet 4.5 (20250929)
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-28 · rating

Claude Sonnet 4.5 (20250929)

1,472.14 rating

Confidence interval: 1,464.31 – 1,479.98

6151 votes

Category
expert
Evaluated configuration
Claude Sonnet 4.5 (20250929)
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-28 · rating

Claude Sonnet 4.5 (20250929)

1,459.46 rating

Confidence interval: 1,444.97 – 1,473.94

1993 votes

Category
french
Evaluated configuration
Claude Sonnet 4.5 (20250929)
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-28 · rating

Claude Sonnet 4.5 (20250929)

1,421.16 rating

Confidence interval: 1,404.72 – 1,437.60

1355 votes

Category
german
Evaluated configuration
Claude Sonnet 4.5 (20250929)
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-28 · rating

Claude Sonnet 4.5 (20250929)

1,461.78 rating

Confidence interval: 1,458.19 – 1,465.37

48280 votes

Category
hard_prompts
Evaluated configuration
Claude Sonnet 4.5 (20250929)
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating

Claude Sonnet 4.5 (20250929)

1,470.10 rating

Confidence interval: 1,465.25 – 1,474.95

21299 votes

Category
hard_prompts_english
Evaluated configuration
Claude Sonnet 4.5 (20250929)
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating

Claude Sonnet 4.5 (20250929)

1,439.94 rating

Confidence interval: 1,434.60 – 1,445.28

15828 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
Claude Sonnet 4.5 (20250929)
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-28 · rating

Claude Sonnet 4.5 (20250929)

1,428.33 rating

Confidence interval: 1,422.90 – 1,433.77

15755 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
Claude Sonnet 4.5 (20250929)
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-28 · rating

Claude Sonnet 4.5 (20250929)

1,450.39 rating

Confidence interval: 1,442.41 – 1,458.36

5983 votes

Category
industry_legal_and_government
Evaluated configuration
Claude Sonnet 4.5 (20250929)
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-28 · rating

Claude Sonnet 4.5 (20250929)

1,447.41 rating

Confidence interval: 1,441.78 – 1,453.04

13597 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
Claude Sonnet 4.5 (20250929)
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-28 · rating

Claude Sonnet 4.5 (20250929)

1,446.36 rating

Confidence interval: 1,437.09 – 1,455.63

4311 votes

Category
industry_mathematical
Evaluated configuration
Claude Sonnet 4.5 (20250929)
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-28 · rating

Claude Sonnet 4.5 (20250929)

1,443.41 rating

Confidence interval: 1,434.85 – 1,451.98

5387 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
Claude Sonnet 4.5 (20250929)
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-28 · rating

Claude Sonnet 4.5 (20250929)

1,472.43 rating

Confidence interval: 1,468.23 – 1,476.64

30303 votes

Category
industry_software_and_it_services
Evaluated configuration
Claude Sonnet 4.5 (20250929)
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-28 · rating

Claude Sonnet 4.5 (20250929)

1,441.79 rating

Confidence interval: 1,436.79 – 1,446.78

18991 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
Claude Sonnet 4.5 (20250929)
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-28 · rating

Claude Sonnet 4.5 (20250929)

1,459.16 rating

Confidence interval: 1,454.57 – 1,463.74

24353 votes

Category
instruction_following
Evaluated configuration
Claude Sonnet 4.5 (20250929)
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-28 · rating

Claude Sonnet 4.5 (20250929)

1,389.82 rating

Confidence interval: 1,368.20 – 1,411.44

785 votes

Category
japanese
Evaluated configuration
Claude Sonnet 4.5 (20250929)
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-10-28 · rating

Claude Sonnet 4.5 (20250929)

1,401.05 rating

Confidence interval: 1,384.07 – 1,418.03

1327 votes

Category
korean
Evaluated configuration
Claude Sonnet 4.5 (20250929)
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-28 · rating

Claude Sonnet 4.5 (20250929)

1,475.24 rating

Confidence interval: 1,470.65 – 1,479.84

26445 votes

Category
longer_query
Evaluated configuration
Claude Sonnet 4.5 (20250929)
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-28 · rating

Claude Sonnet 4.5 (20250929)

1,422.18 rating

Confidence interval: 1,413.82 – 1,430.53

5123 votes

Category
math
Evaluated configuration
Claude Sonnet 4.5 (20250929)
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-28 · rating

Claude Sonnet 4.5 (20250929)

1,464.64 rating

Confidence interval: 1,459.03 – 1,470.24

14471 votes

Category
multi_turn
Evaluated configuration
Claude Sonnet 4.5 (20250929)
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-28 · rating

Claude Sonnet 4.5 (20250929)

1,424.95 rating

Confidence interval: 1,421.41 – 1,428.49

46869 votes

Category
non_english
Evaluated configuration
Claude Sonnet 4.5 (20250929)
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-28 · rating

Claude Sonnet 4.5 (20250929)

1,438.51 rating

Confidence interval: 1,435.66 – 1,441.36

82621 votes

Category
overall
Evaluated configuration
Claude Sonnet 4.5 (20250929)
Evaluation type
human preference
Arena text preference / polishhuman preference · 2026-10-28 · rating

Claude Sonnet 4.5 (20250929)

1,419.05 rating

Confidence interval: 1,406.78 – 1,431.33

2315 votes

Category
polish
Evaluated configuration
Claude Sonnet 4.5 (20250929)
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-28 · rating

Claude Sonnet 4.5 (20250929)

1,436.59 rating

Confidence interval: 1,430.14 – 1,443.03

9572 votes

Category
russian
Evaluated configuration
Claude Sonnet 4.5 (20250929)
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-28 · rating

Claude Sonnet 4.5 (20250929)

1,457.47 rating

Confidence interval: 1,445.10 – 1,469.84

2502 votes

Category
spanish
Evaluated configuration
Claude Sonnet 4.5 (20250929)
Evaluation type
human preference

Run it locally

Inference · memory estimate

Claude Sonnet 4.5 (20250929)

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license
Proprietary