GeekRadar / The comparison desk
COMPARE.
AI models
anthropic

Claude 3.7 Sonnet (20250219)

Model evaluation

Capability & preference

Human preference · Arena

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

Claude 3.7 Sonnet (20250219) 1299.3 Claude 3.7 Sonnet (20250219) · 95% interval 1295.3–1303.3 · 42,566 votes · 2026-11-06

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 · 2026-11-06 · rating

Claude 3.7 Sonnet (20250219)

1,297.43 rating

Confidence interval: 1,285.31 – 1,309.54

2528 votes

Category
chinese
Evaluated configuration
Claude 3.7 Sonnet (20250219)
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-11-06 · rating

Claude 3.7 Sonnet (20250219)

1,338.66 rating

Confidence interval: 1,331.22 – 1,346.10

7073 votes

Category
coding
Evaluated configuration
Claude 3.7 Sonnet (20250219)
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-11-06 · rating

Claude 3.7 Sonnet (20250219)

1,316.61 rating

Confidence interval: 1,308.51 – 1,324.71

5934 votes

Category
creative_writing
Evaluated configuration
Claude 3.7 Sonnet (20250219)
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-11-06 · rating

Claude 3.7 Sonnet (20250219)

1,310.14 rating

Confidence interval: 1,305.13 – 1,315.15

22095 votes

Category
english
Evaluated configuration
Claude 3.7 Sonnet (20250219)
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-11-06 · rating

Claude 3.7 Sonnet (20250219)

1,240.49 rating

Confidence interval: 1,235.16 – 1,245.82

31236 votes

Category
exclude_ties
Evaluated configuration
Claude 3.7 Sonnet (20250219)
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-11-06 · rating

Claude 3.7 Sonnet (20250219)

1,300.52 rating

Confidence interval: 1,287.53 – 1,313.52

2147 votes

Category
expert
Evaluated configuration
Claude 3.7 Sonnet (20250219)
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-11-06 · rating

Claude 3.7 Sonnet (20250219)

1,302.79 rating

Confidence interval: 1,276.24 – 1,329.35

551 votes

Category
french
Evaluated configuration
Claude 3.7 Sonnet (20250219)
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-11-06 · rating

Claude 3.7 Sonnet (20250219)

1,280.74 rating

Confidence interval: 1,261.76 – 1,299.73

1084 votes

Category
german
Evaluated configuration
Claude 3.7 Sonnet (20250219)
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-11-06 · rating

Claude 3.7 Sonnet (20250219)

1,314.96 rating

Confidence interval: 1,309.34 – 1,320.58

15170 votes

Category
hard_prompts
Evaluated configuration
Claude 3.7 Sonnet (20250219)
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-11-06 · rating

Claude 3.7 Sonnet (20250219)

1,320.07 rating

Confidence interval: 1,312.92 – 1,327.22

7982 votes

Category
hard_prompts_english
Evaluated configuration
Claude 3.7 Sonnet (20250219)
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-11-06 · rating

Claude 3.7 Sonnet (20250219)

1,272.67 rating

Confidence interval: 1,264.24 – 1,281.11

5753 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
Claude 3.7 Sonnet (20250219)
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-11-06 · rating

Claude 3.7 Sonnet (20250219)

1,300.18 rating

Confidence interval: 1,292.84 – 1,307.51

7332 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
Claude 3.7 Sonnet (20250219)
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-11-06 · rating

Claude 3.7 Sonnet (20250219)

1,300.19 rating

Confidence interval: 1,287.67 – 1,312.71

2476 votes

Category
industry_legal_and_government
Evaluated configuration
Claude 3.7 Sonnet (20250219)
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-11-06 · rating

Claude 3.7 Sonnet (20250219)

1,285.00 rating

Confidence interval: 1,277.34 – 1,292.66

7444 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
Claude 3.7 Sonnet (20250219)
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-11-06 · rating

Claude 3.7 Sonnet (20250219)

1,313.76 rating

Confidence interval: 1,303.41 – 1,324.12

3177 votes

Category
industry_mathematical
Evaluated configuration
Claude 3.7 Sonnet (20250219)
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-11-06 · rating

Claude 3.7 Sonnet (20250219)

1,258.18 rating

Confidence interval: 1,244.45 – 1,271.90

2255 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
Claude 3.7 Sonnet (20250219)
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-11-06 · rating

Claude 3.7 Sonnet (20250219)

1,314.12 rating

Confidence interval: 1,308.08 – 1,320.15

12735 votes

Category
industry_software_and_it_services
Evaluated configuration
Claude 3.7 Sonnet (20250219)
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-11-06 · rating

Claude 3.7 Sonnet (20250219)

1,323.99 rating

Confidence interval: 1,317.43 – 1,330.54

9711 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
Claude 3.7 Sonnet (20250219)
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-11-06 · rating

Claude 3.7 Sonnet (20250219)

1,324.39 rating

Confidence interval: 1,318.49 – 1,330.30

12305 votes

Category
instruction_following
Evaluated configuration
Claude 3.7 Sonnet (20250219)
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-11-06 · rating

Claude 3.7 Sonnet (20250219)

1,260.29 rating

Confidence interval: 1,240.97 – 1,279.60

1009 votes

Category
japanese
Evaluated configuration
Claude 3.7 Sonnet (20250219)
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-11-06 · rating

Claude 3.7 Sonnet (20250219)

1,239.48 rating

Confidence interval: 1,215.82 – 1,263.15

702 votes

Category
korean
Evaluated configuration
Claude 3.7 Sonnet (20250219)
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-11-06 · rating

Claude 3.7 Sonnet (20250219)

1,354.61 rating

Confidence interval: 1,347.11 – 1,362.10

6664 votes

Category
longer_query
Evaluated configuration
Claude 3.7 Sonnet (20250219)
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-11-06 · rating

Claude 3.7 Sonnet (20250219)

1,318.56 rating

Confidence interval: 1,308.60 – 1,328.53

3352 votes

Category
math
Evaluated configuration
Claude 3.7 Sonnet (20250219)
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-11-06 · rating

Claude 3.7 Sonnet (20250219)

1,339.12 rating

Confidence interval: 1,331.43 – 1,346.80

6826 votes

Category
multi_turn
Evaluated configuration
Claude 3.7 Sonnet (20250219)
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-11-06 · rating

Claude 3.7 Sonnet (20250219)

1,286.11 rating

Confidence interval: 1,281.04 – 1,291.19

20460 votes

Category
non_english
Evaluated configuration
Claude 3.7 Sonnet (20250219)
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-11-06 · rating

Claude 3.7 Sonnet (20250219)

1,299.32 rating

Confidence interval: 1,295.33 – 1,303.30

42566 votes

Category
overall
Evaluated configuration
Claude 3.7 Sonnet (20250219)
Evaluation type
human preference
Arena text preference / polishhuman preference · 2026-11-06 · rating

Claude 3.7 Sonnet (20250219)

1,278.41 rating

Confidence interval: 1,264.43 – 1,292.39

2175 votes

Category
polish
Evaluated configuration
Claude 3.7 Sonnet (20250219)
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-11-06 · rating

Claude 3.7 Sonnet (20250219)

1,310.24 rating

Confidence interval: 1,300.20 – 1,320.27

3506 votes

Category
russian
Evaluated configuration
Claude 3.7 Sonnet (20250219)
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-11-06 · rating

Claude 3.7 Sonnet (20250219)

1,289.75 rating

Confidence interval: 1,265.13 – 1,314.37

623 votes

Category
spanish
Evaluated configuration
Claude 3.7 Sonnet (20250219)
Evaluation type
human preference

Run it locally

Inference · memory estimate

Claude 3.7 Sonnet (20250219)

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license
Proprietary