GeekRadar / The comparison desk
COMPARE.
AI models
anthropic

Claude Sonnet 4.5 (High, 32K, 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 (High, 32K, 20250929) 1434.3 Claude Sonnet 4.5 (High, 32K, 20250929) · 95% interval 1431.6–1437.1 · 83,310 votes · 2026-11-11

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 · 2026-11-11 · rating

Claude Sonnet 4.5 (High, 32K, 20250929)

1,458.49 rating

Confidence interval: 1,450.40 – 1,466.57

5832 votes

Category
chinese
Evaluated configuration
Claude Sonnet 4.5 (High, 32K, 20250929)
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-11-11 · rating

Claude Sonnet 4.5 (High, 32K, 20250929)

1,488.11 rating

Confidence interval: 1,483.26 – 1,492.95

19781 votes

Category
coding
Evaluated configuration
Claude Sonnet 4.5 (High, 32K, 20250929)
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-11-11 · rating

Claude Sonnet 4.5 (High, 32K, 20250929)

1,424.32 rating

Confidence interval: 1,418.43 – 1,430.21

12558 votes

Category
creative_writing
Evaluated configuration
Claude Sonnet 4.5 (High, 32K, 20250929)
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-11-11 · rating

Claude Sonnet 4.5 (High, 32K, 20250929)

1,448.37 rating

Confidence interval: 1,444.53 – 1,452.22

35928 votes

Category
english
Evaluated configuration
Claude Sonnet 4.5 (High, 32K, 20250929)
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-11-11 · rating

Claude Sonnet 4.5 (High, 32K, 20250929)

1,427.05 rating

Confidence interval: 1,423.32 – 1,430.78

61267 votes

Category
exclude_ties
Evaluated configuration
Claude Sonnet 4.5 (High, 32K, 20250929)
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-11-11 · rating

Claude Sonnet 4.5 (High, 32K, 20250929)

1,482.54 rating

Confidence interval: 1,474.59 – 1,490.49

6014 votes

Category
expert
Evaluated configuration
Claude Sonnet 4.5 (High, 32K, 20250929)
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-11-11 · rating

Claude Sonnet 4.5 (High, 32K, 20250929)

1,451.41 rating

Confidence interval: 1,436.94 – 1,465.88

1964 votes

Category
french
Evaluated configuration
Claude Sonnet 4.5 (High, 32K, 20250929)
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-11-11 · rating

Claude Sonnet 4.5 (High, 32K, 20250929)

1,426.67 rating

Confidence interval: 1,410.06 – 1,443.28

1352 votes

Category
german
Evaluated configuration
Claude Sonnet 4.5 (High, 32K, 20250929)
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-11-11 · rating

Claude Sonnet 4.5 (High, 32K, 20250929)

1,461.49 rating

Confidence interval: 1,457.96 – 1,465.02

48089 votes

Category
hard_prompts
Evaluated configuration
Claude Sonnet 4.5 (High, 32K, 20250929)
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-11-11 · rating

Claude Sonnet 4.5 (High, 32K, 20250929)

1,471.09 rating

Confidence interval: 1,466.32 – 1,475.87

20968 votes

Category
hard_prompts_english
Evaluated configuration
Claude Sonnet 4.5 (High, 32K, 20250929)
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-11-11 · rating

Claude Sonnet 4.5 (High, 32K, 20250929)

1,432.66 rating

Confidence interval: 1,427.36 – 1,437.96

15956 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
Claude Sonnet 4.5 (High, 32K, 20250929)
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-11-11 · rating

Claude Sonnet 4.5 (High, 32K, 20250929)

1,419.93 rating

Confidence interval: 1,414.64 – 1,425.22

16010 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
Claude Sonnet 4.5 (High, 32K, 20250929)
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-11-11 · rating

Claude Sonnet 4.5 (High, 32K, 20250929)

1,441.74 rating

Confidence interval: 1,433.67 – 1,449.81

5815 votes

Category
industry_legal_and_government
Evaluated configuration
Claude Sonnet 4.5 (High, 32K, 20250929)
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-11-11 · rating

Claude Sonnet 4.5 (High, 32K, 20250929)

1,441.00 rating

Confidence interval: 1,435.44 – 1,446.56

13505 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
Claude Sonnet 4.5 (High, 32K, 20250929)
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-11-11 · rating

Claude Sonnet 4.5 (High, 32K, 20250929)

1,462.58 rating

Confidence interval: 1,453.24 – 1,471.92

4145 votes

Category
industry_mathematical
Evaluated configuration
Claude Sonnet 4.5 (High, 32K, 20250929)
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-11-11 · rating

Claude Sonnet 4.5 (High, 32K, 20250929)

1,429.97 rating

Confidence interval: 1,421.45 – 1,438.50

5342 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
Claude Sonnet 4.5 (High, 32K, 20250929)
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-11-11 · rating

Claude Sonnet 4.5 (High, 32K, 20250929)

1,470.69 rating

Confidence interval: 1,466.57 – 1,474.81

30188 votes

Category
industry_software_and_it_services
Evaluated configuration
Claude Sonnet 4.5 (High, 32K, 20250929)
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-11-11 · rating

Claude Sonnet 4.5 (High, 32K, 20250929)

1,431.63 rating

Confidence interval: 1,426.73 – 1,436.54

19196 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
Claude Sonnet 4.5 (High, 32K, 20250929)
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-11-11 · rating

Claude Sonnet 4.5 (High, 32K, 20250929)

1,457.73 rating

Confidence interval: 1,453.24 – 1,462.23

24464 votes

Category
instruction_following
Evaluated configuration
Claude Sonnet 4.5 (High, 32K, 20250929)
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-11-11 · rating

Claude Sonnet 4.5 (High, 32K, 20250929)

1,372.94 rating

Confidence interval: 1,351.19 – 1,394.70

770 votes

Category
japanese
Evaluated configuration
Claude Sonnet 4.5 (High, 32K, 20250929)
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-11-11 · rating

Claude Sonnet 4.5 (High, 32K, 20250929)

1,367.51 rating

Confidence interval: 1,350.56 – 1,384.46

1353 votes

Category
korean
Evaluated configuration
Claude Sonnet 4.5 (High, 32K, 20250929)
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-11-11 · rating

Claude Sonnet 4.5 (High, 32K, 20250929)

1,474.72 rating

Confidence interval: 1,470.17 – 1,479.26

25835 votes

Category
longer_query
Evaluated configuration
Claude Sonnet 4.5 (High, 32K, 20250929)
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-11-11 · rating

Claude Sonnet 4.5 (High, 32K, 20250929)

1,449.08 rating

Confidence interval: 1,440.65 – 1,457.52

5013 votes

Category
math
Evaluated configuration
Claude Sonnet 4.5 (High, 32K, 20250929)
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-11-11 · rating

Claude Sonnet 4.5 (High, 32K, 20250929)

1,453.34 rating

Confidence interval: 1,447.78 – 1,458.89

14238 votes

Category
multi_turn
Evaluated configuration
Claude Sonnet 4.5 (High, 32K, 20250929)
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-11-11 · rating

Claude Sonnet 4.5 (High, 32K, 20250929)

1,417.76 rating

Confidence interval: 1,414.31 – 1,421.21

47348 votes

Category
non_english
Evaluated configuration
Claude Sonnet 4.5 (High, 32K, 20250929)
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-11-11 · rating

Claude Sonnet 4.5 (High, 32K, 20250929)

1,434.35 rating

Confidence interval: 1,431.59 – 1,437.11

83310 votes

Category
overall
Evaluated configuration
Claude Sonnet 4.5 (High, 32K, 20250929)
Evaluation type
human preference
Arena text preference / polishhuman preference · 2026-11-11 · rating

Claude Sonnet 4.5 (High, 32K, 20250929)

1,422.79 rating

Confidence interval: 1,410.14 – 1,435.43

2217 votes

Category
polish
Evaluated configuration
Claude Sonnet 4.5 (High, 32K, 20250929)
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-11-11 · rating

Claude Sonnet 4.5 (High, 32K, 20250929)

1,428.47 rating

Confidence interval: 1,422.07 – 1,434.88

9466 votes

Category
russian
Evaluated configuration
Claude Sonnet 4.5 (High, 32K, 20250929)
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-11-11 · rating

Claude Sonnet 4.5 (High, 32K, 20250929)

1,442.21 rating

Confidence interval: 1,429.74 – 1,454.68

2549 votes

Category
spanish
Evaluated configuration
Claude Sonnet 4.5 (High, 32K, 20250929)
Evaluation type
human preference

Run it locally

Inference · memory estimate

Claude Sonnet 4.5 (High, 32K, 20250929)

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license
Proprietary