GeekRadar / The comparison desk
COMPARE.
AI models
anthropic

claude-3-opus-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-opus-20240229 1262.0 claude-3-opus-20240229 · 95% interval 1258.9–1265.0 · 194,909 votes · 2026-10-15

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

claude-3-opus-20240229

1,247.75 rating

Confidence interval: 1,239.75 – 1,255.75

23348 votes

Category
chinese
Evaluated configuration
claude-3-opus-20240229
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-15 · rating

claude-3-opus-20240229

1,264.00 rating

Confidence interval: 1,258.16 – 1,269.85

33748 votes

Category
coding
Evaluated configuration
claude-3-opus-20240229
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating

claude-3-opus-20240229

1,234.66 rating

Confidence interval: 1,228.16 – 1,241.16

28395 votes

Category
creative_writing
Evaluated configuration
claude-3-opus-20240229
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating

claude-3-opus-20240229

1,262.76 rating

Confidence interval: 1,258.78 – 1,266.75

108426 votes

Category
english
Evaluated configuration
claude-3-opus-20240229
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating

claude-3-opus-20240229

1,174.70 rating

Confidence interval: 1,169.98 – 1,179.41

133332 votes

Category
exclude_ties
Evaluated configuration
claude-3-opus-20240229
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-15 · rating

claude-3-opus-20240229

1,222.47 rating

Confidence interval: 1,213.16 – 1,231.78

10374 votes

Category
expert
Evaluated configuration
claude-3-opus-20240229
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-15 · rating

claude-3-opus-20240229

1,275.17 rating

Confidence interval: 1,256.30 – 1,294.03

2847 votes

Category
french
Evaluated configuration
claude-3-opus-20240229
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-15 · rating

claude-3-opus-20240229

1,258.37 rating

Confidence interval: 1,245.67 – 1,271.07

5672 votes

Category
german
Evaluated configuration
claude-3-opus-20240229
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating

claude-3-opus-20240229

1,245.30 rating

Confidence interval: 1,240.50 – 1,250.09

55074 votes

Category
hard_prompts
Evaluated configuration
claude-3-opus-20240229
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating

claude-3-opus-20240229

1,243.13 rating

Confidence interval: 1,237.34 – 1,248.92

34293 votes

Category
hard_prompts_english
Evaluated configuration
claude-3-opus-20240229
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating

claude-3-opus-20240229

1,222.06 rating

Confidence interval: 1,215.11 – 1,229.01

21697 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
claude-3-opus-20240229
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating

claude-3-opus-20240229

1,220.11 rating

Confidence interval: 1,214.04 – 1,226.17

29359 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
claude-3-opus-20240229
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-15 · rating

claude-3-opus-20240229

1,264.31 rating

Confidence interval: 1,254.91 – 1,273.71

11095 votes

Category
industry_legal_and_government
Evaluated configuration
claude-3-opus-20240229
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating

claude-3-opus-20240229

1,253.81 rating

Confidence interval: 1,247.71 – 1,259.91

33790 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
claude-3-opus-20240229
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-15 · rating

claude-3-opus-20240229

1,263.77 rating

Confidence interval: 1,256.99 – 1,270.54

22970 votes

Category
industry_mathematical
Evaluated configuration
claude-3-opus-20240229
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating

claude-3-opus-20240229

1,231.14 rating

Confidence interval: 1,221.28 – 1,240.99

10327 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
claude-3-opus-20240229
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating

claude-3-opus-20240229

1,258.80 rating

Confidence interval: 1,253.80 – 1,263.79

54971 votes

Category
industry_software_and_it_services
Evaluated configuration
claude-3-opus-20240229
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating

claude-3-opus-20240229

1,268.61 rating

Confidence interval: 1,263.36 – 1,273.86

49929 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
claude-3-opus-20240229
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating

claude-3-opus-20240229

1,248.14 rating

Confidence interval: 1,243.66 – 1,252.61

72001 votes

Category
instruction_following
Evaluated configuration
claude-3-opus-20240229
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-15 · rating

claude-3-opus-20240229

1,203.94 rating

Confidence interval: 1,189.57 – 1,218.32

3650 votes

Category
japanese
Evaluated configuration
claude-3-opus-20240229
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-10-15 · rating

claude-3-opus-20240229

1,187.39 rating

Confidence interval: 1,170.32 – 1,204.46

3906 votes

Category
korean
Evaluated configuration
claude-3-opus-20240229
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-15 · rating

claude-3-opus-20240229

1,258.81 rating

Confidence interval: 1,252.44 – 1,265.18

23374 votes

Category
longer_query
Evaluated configuration
claude-3-opus-20240229
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-15 · rating

claude-3-opus-20240229

1,272.49 rating

Confidence interval: 1,266.02 – 1,278.95

25769 votes

Category
math
Evaluated configuration
claude-3-opus-20240229
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-15 · rating

claude-3-opus-20240229

1,275.32 rating

Confidence interval: 1,269.18 – 1,281.46

31134 votes

Category
multi_turn
Evaluated configuration
claude-3-opus-20240229
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating

claude-3-opus-20240229

1,257.64 rating

Confidence interval: 1,253.48 – 1,261.81

86483 votes

Category
non_english
Evaluated configuration
claude-3-opus-20240229
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating

claude-3-opus-20240229

1,261.97 rating

Confidence interval: 1,258.93 – 1,265.00

194909 votes

Category
overall
Evaluated configuration
claude-3-opus-20240229
Evaluation type
human preference
Arena text preference / polishhuman preference · 2026-10-15 · rating

claude-3-opus-20240229

1,351.93 rating

Confidence interval: 1,303.75 – 1,400.10

371 votes

Category
polish
Evaluated configuration
claude-3-opus-20240229
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-15 · rating

claude-3-opus-20240229

1,279.97 rating

Confidence interval: 1,273.08 – 1,286.86

21089 votes

Category
russian
Evaluated configuration
claude-3-opus-20240229
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-15 · rating

claude-3-opus-20240229

1,246.21 rating

Confidence interval: 1,227.15 – 1,265.28

2569 votes

Category
spanish
Evaluated configuration
claude-3-opus-20240229
Evaluation type
human preference

Run it locally

Inference · memory estimate

claude-3-opus-20240229

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license
Proprietary