GeekRadar / The comparison desk
COMPARE.
AI models
anthropic

Claude 3 Sonnet (20240229)

Model evaluation

Capability & preference

Human preference · Arena

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

Claude 3 Sonnet (20240229) 1218.0 Claude 3 Sonnet (20240229) · 95% interval 1214.1–1222.0 · 109,284 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 3 Sonnet (20240229)

1,188.56 rating

Confidence interval: 1,179.29 – 1,197.83

15648 votes

Category
chinese
Evaluated configuration
Claude 3 Sonnet (20240229)
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-28 · rating

Claude 3 Sonnet (20240229)

1,222.44 rating

Confidence interval: 1,215.16 – 1,229.72

18888 votes

Category
coding
Evaluated configuration
Claude 3 Sonnet (20240229)
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-28 · rating

Claude 3 Sonnet (20240229)

1,185.88 rating

Confidence interval: 1,177.66 – 1,194.09

15592 votes

Category
creative_writing
Evaluated configuration
Claude 3 Sonnet (20240229)
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-28 · rating

Claude 3 Sonnet (20240229)

1,226.10 rating

Confidence interval: 1,221.21 – 1,230.99

60970 votes

Category
english
Evaluated configuration
Claude 3 Sonnet (20240229)
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating

Claude 3 Sonnet (20240229)

1,108.14 rating

Confidence interval: 1,102.34 – 1,113.94

73557 votes

Category
exclude_ties
Evaluated configuration
Claude 3 Sonnet (20240229)
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-28 · rating

Claude 3 Sonnet (20240229)

1,172.24 rating

Confidence interval: 1,160.46 – 1,184.02

5614 votes

Category
expert
Evaluated configuration
Claude 3 Sonnet (20240229)
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-28 · rating

Claude 3 Sonnet (20240229)

1,229.43 rating

Confidence interval: 1,208.48 – 1,250.39

1824 votes

Category
french
Evaluated configuration
Claude 3 Sonnet (20240229)
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-28 · rating

Claude 3 Sonnet (20240229)

1,203.73 rating

Confidence interval: 1,188.51 – 1,218.95

3097 votes

Category
german
Evaluated configuration
Claude 3 Sonnet (20240229)
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-28 · rating

Claude 3 Sonnet (20240229)

1,196.88 rating

Confidence interval: 1,190.97 – 1,202.78

30725 votes

Category
hard_prompts
Evaluated configuration
Claude 3 Sonnet (20240229)
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating

Claude 3 Sonnet (20240229)

1,202.41 rating

Confidence interval: 1,195.21 – 1,209.61

19193 votes

Category
hard_prompts_english
Evaluated configuration
Claude 3 Sonnet (20240229)
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating

Claude 3 Sonnet (20240229)

1,173.95 rating

Confidence interval: 1,165.02 – 1,182.88

11853 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
Claude 3 Sonnet (20240229)
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-28 · rating

Claude 3 Sonnet (20240229)

1,171.29 rating

Confidence interval: 1,163.47 – 1,179.11

15961 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
Claude 3 Sonnet (20240229)
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-28 · rating

Claude 3 Sonnet (20240229)

1,213.84 rating

Confidence interval: 1,202.01 – 1,225.68

6062 votes

Category
industry_legal_and_government
Evaluated configuration
Claude 3 Sonnet (20240229)
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-28 · rating

Claude 3 Sonnet (20240229)

1,209.11 rating

Confidence interval: 1,201.49 – 1,216.74

19089 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
Claude 3 Sonnet (20240229)
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-28 · rating

Claude 3 Sonnet (20240229)

1,213.02 rating

Confidence interval: 1,204.61 – 1,221.42

12413 votes

Category
industry_mathematical
Evaluated configuration
Claude 3 Sonnet (20240229)
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-28 · rating

Claude 3 Sonnet (20240229)

1,188.13 rating

Confidence interval: 1,176.06 – 1,200.21

5941 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
Claude 3 Sonnet (20240229)
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-28 · rating

Claude 3 Sonnet (20240229)

1,220.75 rating

Confidence interval: 1,214.62 – 1,226.88

31448 votes

Category
industry_software_and_it_services
Evaluated configuration
Claude 3 Sonnet (20240229)
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-28 · rating

Claude 3 Sonnet (20240229)

1,217.65 rating

Confidence interval: 1,211.07 – 1,224.24

27297 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
Claude 3 Sonnet (20240229)
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-28 · rating

Claude 3 Sonnet (20240229)

1,199.22 rating

Confidence interval: 1,193.62 – 1,204.82

38802 votes

Category
instruction_following
Evaluated configuration
Claude 3 Sonnet (20240229)
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-28 · rating

Claude 3 Sonnet (20240229)

1,131.27 rating

Confidence interval: 1,112.98 – 1,149.57

1753 votes

Category
japanese
Evaluated configuration
Claude 3 Sonnet (20240229)
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-10-28 · rating

Claude 3 Sonnet (20240229)

1,127.73 rating

Confidence interval: 1,108.27 – 1,147.19

2059 votes

Category
korean
Evaluated configuration
Claude 3 Sonnet (20240229)
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-28 · rating

Claude 3 Sonnet (20240229)

1,210.74 rating

Confidence interval: 1,202.14 – 1,219.35

11870 votes

Category
longer_query
Evaluated configuration
Claude 3 Sonnet (20240229)
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-28 · rating

Claude 3 Sonnet (20240229)

1,213.39 rating

Confidence interval: 1,205.35 – 1,221.44

13766 votes

Category
math
Evaluated configuration
Claude 3 Sonnet (20240229)
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-28 · rating

Claude 3 Sonnet (20240229)

1,226.85 rating

Confidence interval: 1,218.92 – 1,234.77

17062 votes

Category
multi_turn
Evaluated configuration
Claude 3 Sonnet (20240229)
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-28 · rating

Claude 3 Sonnet (20240229)

1,205.03 rating

Confidence interval: 1,199.82 – 1,210.24

48314 votes

Category
non_english
Evaluated configuration
Claude 3 Sonnet (20240229)
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-28 · rating

Claude 3 Sonnet (20240229)

1,218.01 rating

Confidence interval: 1,214.06 – 1,221.97

109284 votes

Category
overall
Evaluated configuration
Claude 3 Sonnet (20240229)
Evaluation type
human preference
Arena text preference / polishhuman preference · 2026-10-28 · rating

Claude 3 Sonnet (20240229)

1,259.29 rating

Confidence interval: 1,207.00 – 1,311.57

284 votes

Category
polish
Evaluated configuration
Claude 3 Sonnet (20240229)
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-28 · rating

Claude 3 Sonnet (20240229)

1,226.42 rating

Confidence interval: 1,217.32 – 1,235.52

10794 votes

Category
russian
Evaluated configuration
Claude 3 Sonnet (20240229)
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-28 · rating

Claude 3 Sonnet (20240229)

1,204.26 rating

Confidence interval: 1,182.34 – 1,226.18

1448 votes

Category
spanish
Evaluated configuration
Claude 3 Sonnet (20240229)
Evaluation type
human preference

Run it locally

Inference · memory estimate

Claude 3 Sonnet (20240229)

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license
Proprietary