GeekRadar / The comparison desk
COMPARE.
AI models
anthropic

Claude Sonnet 3.7

AI model Released

Context window
200,000 tokens
Maximum output
64,000 tokens
Open weights
No

Capability & preference

Human preference · Arena

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

Claude Sonnet 3.7 1299.4 claude-3-7-sonnet-20250219 · 95% interval 1295.4–1303.4 · 42,572 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 Sonnet 3.7

1,298.77 rating

Confidence interval: 1,286.03 – 1,311.52

2311 votes

Category
chinese
Evaluated configuration
claude-3-7-sonnet-20250219-thinking-32k
Evaluation type
human preference

1,297.32 rating

Confidence interval: 1,285.21 – 1,309.43

2526 votes

Category
chinese
Evaluated configuration
claude-3-7-sonnet-20250219
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-15 · rating

Claude Sonnet 3.7

1,360.33 rating

Confidence interval: 1,352.41 – 1,368.25

6135 votes

Category
coding
Evaluated configuration
claude-3-7-sonnet-20250219-thinking-32k
Evaluation type
human preference

1,338.94 rating

Confidence interval: 1,331.50 – 1,346.37

7071 votes

Category
coding
Evaluated configuration
claude-3-7-sonnet-20250219
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating

Claude Sonnet 3.7

1,331.72 rating

Confidence interval: 1,323.09 – 1,340.35

5146 votes

Category
creative_writing
Evaluated configuration
claude-3-7-sonnet-20250219-thinking-32k
Evaluation type
human preference

1,316.71 rating

Confidence interval: 1,308.61 – 1,324.80

5936 votes

Category
creative_writing
Evaluated configuration
claude-3-7-sonnet-20250219
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating

Claude Sonnet 3.7

1,326.34 rating

Confidence interval: 1,321.03 – 1,331.65

18910 votes

Category
english
Evaluated configuration
claude-3-7-sonnet-20250219-thinking-32k
Evaluation type
human preference

1,310.16 rating

Confidence interval: 1,305.15 – 1,315.16

22100 votes

Category
english
Evaluated configuration
claude-3-7-sonnet-20250219
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating

Claude Sonnet 3.7

1,260.79 rating

Confidence interval: 1,255.19 – 1,266.40

28180 votes

Category
exclude_ties
Evaluated configuration
claude-3-7-sonnet-20250219-thinking-32k
Evaluation type
human preference

1,240.62 rating

Confidence interval: 1,235.29 – 1,245.94

31238 votes

Category
exclude_ties
Evaluated configuration
claude-3-7-sonnet-20250219
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-15 · rating

Claude Sonnet 3.7

1,320.48 rating

Confidence interval: 1,306.59 – 1,334.36

1889 votes

Category
expert
Evaluated configuration
claude-3-7-sonnet-20250219-thinking-32k
Evaluation type
human preference

1,300.66 rating

Confidence interval: 1,287.67 – 1,313.65

2146 votes

Category
expert
Evaluated configuration
claude-3-7-sonnet-20250219
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-15 · rating

Claude Sonnet 3.7

1,302.77 rating

Confidence interval: 1,276.27 – 1,329.27

552 votes

Category
french
Evaluated configuration
claude-3-7-sonnet-20250219
Evaluation type
human preference

1,288.25 rating

Confidence interval: 1,260.28 – 1,316.23

514 votes

Category
french
Evaluated configuration
claude-3-7-sonnet-20250219-thinking-32k
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-15 · rating

Claude Sonnet 3.7

1,300.77 rating

Confidence interval: 1,281.44 – 1,320.10

1022 votes

Category
german
Evaluated configuration
claude-3-7-sonnet-20250219-thinking-32k
Evaluation type
human preference

1,281.02 rating

Confidence interval: 1,262.03 – 1,300.00

1082 votes

Category
german
Evaluated configuration
claude-3-7-sonnet-20250219
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating

Claude Sonnet 3.7

1,332.94 rating

Confidence interval: 1,327.02 – 1,338.86

13726 votes

Category
hard_prompts
Evaluated configuration
claude-3-7-sonnet-20250219-thinking-32k
Evaluation type
human preference

1,315.09 rating

Confidence interval: 1,309.48 – 1,320.71

15169 votes

Category
hard_prompts
Evaluated configuration
claude-3-7-sonnet-20250219
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating

Claude Sonnet 3.7

1,339.73 rating

Confidence interval: 1,332.14 – 1,347.31

6984 votes

Category
hard_prompts_english
Evaluated configuration
claude-3-7-sonnet-20250219-thinking-32k
Evaluation type
human preference

1,320.14 rating

Confidence interval: 1,313.00 – 1,327.29

7984 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 · 2026-10-15 · rating

Claude Sonnet 3.7

1,272.91 rating

Confidence interval: 1,264.48 – 1,281.34

5755 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
claude-3-7-sonnet-20250219
Evaluation type
human preference

1,269.07 rating

Confidence interval: 1,260.50 – 1,277.65

5593 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
claude-3-7-sonnet-20250219-thinking-32k
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating

Claude Sonnet 3.7

1,313.29 rating

Confidence interval: 1,305.58 – 1,320.99

6543 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
claude-3-7-sonnet-20250219-thinking-32k
Evaluation type
human preference

1,300.20 rating

Confidence interval: 1,292.87 – 1,307.53

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 · 2026-10-15 · rating

Claude Sonnet 3.7

1,301.26 rating

Confidence interval: 1,288.87 – 1,313.65

2474 votes

Category
industry_legal_and_government
Evaluated configuration
claude-3-7-sonnet-20250219-thinking-32k
Evaluation type
human preference

1,299.82 rating

Confidence interval: 1,287.31 – 1,312.34

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 · 2026-10-15 · rating

Claude Sonnet 3.7

1,298.55 rating

Confidence interval: 1,290.50 – 1,306.59

6706 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
claude-3-7-sonnet-20250219-thinking-32k
Evaluation type
human preference

1,284.97 rating

Confidence interval: 1,277.32 – 1,292.62

7446 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 · 2026-10-15 · rating

Claude Sonnet 3.7

1,338.79 rating

Confidence interval: 1,327.30 – 1,350.29

2547 votes

Category
industry_mathematical
Evaluated configuration
claude-3-7-sonnet-20250219-thinking-32k
Evaluation type
human preference

1,313.64 rating

Confidence interval: 1,303.29 – 1,323.99

3177 votes

Category
industry_mathematical
Evaluated configuration
claude-3-7-sonnet-20250219
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating

Claude Sonnet 3.7

1,287.93 rating

Confidence interval: 1,274.50 – 1,301.36

2242 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
claude-3-7-sonnet-20250219-thinking-32k
Evaluation type
human preference

1,258.29 rating

Confidence interval: 1,244.58 – 1,272.00

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 · 2026-10-15 · rating

Claude Sonnet 3.7

1,323.83 rating

Confidence interval: 1,317.44 – 1,330.22

11412 votes

Category
industry_software_and_it_services
Evaluated configuration
claude-3-7-sonnet-20250219-thinking-32k
Evaluation type
human preference

1,314.26 rating

Confidence interval: 1,308.24 – 1,320.29

12734 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 · 2026-10-15 · rating

Claude Sonnet 3.7

1,337.69 rating

Confidence interval: 1,330.84 – 1,344.54

8764 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
claude-3-7-sonnet-20250219-thinking-32k
Evaluation type
human preference

1,324.04 rating

Confidence interval: 1,317.49 – 1,330.59

9714 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 · 2026-10-15 · rating

Claude Sonnet 3.7

1,351.67 rating

Confidence interval: 1,345.24 – 1,358.09

10122 votes

Category
instruction_following
Evaluated configuration
claude-3-7-sonnet-20250219-thinking-32k
Evaluation type
human preference

1,324.46 rating

Confidence interval: 1,318.56 – 1,330.36

12306 votes

Category
instruction_following
Evaluated configuration
claude-3-7-sonnet-20250219
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-15 · rating

Claude Sonnet 3.7

1,266.50 rating

Confidence interval: 1,246.86 – 1,286.14

1019 votes

Category
japanese
Evaluated configuration
claude-3-7-sonnet-20250219-thinking-32k
Evaluation type
human preference

1,260.20 rating

Confidence interval: 1,240.94 – 1,279.47

1009 votes

Category
japanese
Evaluated configuration
claude-3-7-sonnet-20250219
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-10-15 · rating

Claude Sonnet 3.7

1,249.26 rating

Confidence interval: 1,225.95 – 1,272.56

767 votes

Category
korean
Evaluated configuration
claude-3-7-sonnet-20250219-thinking-32k
Evaluation type
human preference

1,239.32 rating

Confidence interval: 1,215.68 – 1,262.97

702 votes

Category
korean
Evaluated configuration
claude-3-7-sonnet-20250219
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-15 · rating

Claude Sonnet 3.7

1,372.71 rating

Confidence interval: 1,364.82 – 1,380.59

6136 votes

Category
longer_query
Evaluated configuration
claude-3-7-sonnet-20250219-thinking-32k
Evaluation type
human preference

1,354.59 rating

Confidence interval: 1,347.10 – 1,362.08

6667 votes

Category
longer_query
Evaluated configuration
claude-3-7-sonnet-20250219
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-15 · rating

Claude Sonnet 3.7

1,336.30 rating

Confidence interval: 1,325.33 – 1,347.26

2788 votes

Category
math
Evaluated configuration
claude-3-7-sonnet-20250219-thinking-32k
Evaluation type
human preference

1,318.56 rating

Confidence interval: 1,308.61 – 1,328.52

3352 votes

Category
math
Evaluated configuration
claude-3-7-sonnet-20250219
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-15 · rating

Claude Sonnet 3.7

1,339.33 rating

Confidence interval: 1,331.65 – 1,347.01

6826 votes

Category
multi_turn
Evaluated configuration
claude-3-7-sonnet-20250219
Evaluation type
human preference

1,337.71 rating

Confidence interval: 1,329.61 – 1,345.82

6143 votes

Category
multi_turn
Evaluated configuration
claude-3-7-sonnet-20250219-thinking-32k
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating

Claude Sonnet 3.7

1,296.04 rating

Confidence interval: 1,290.73 – 1,301.35

19237 votes

Category
non_english
Evaluated configuration
claude-3-7-sonnet-20250219-thinking-32k
Evaluation type
human preference

1,286.27 rating

Confidence interval: 1,281.20 – 1,291.34

20461 votes

Category
non_english
Evaluated configuration
claude-3-7-sonnet-20250219
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating

Claude Sonnet 3.7

1,314.32 rating

Confidence interval: 1,310.08 – 1,318.56

38159 votes

Category
overall
Evaluated configuration
claude-3-7-sonnet-20250219-thinking-32k
Evaluation type
human preference

1,299.39 rating

Confidence interval: 1,295.40 – 1,303.37

42572 votes

Category
overall
Evaluated configuration
claude-3-7-sonnet-20250219
Evaluation type
human preference
Arena text preference / polishhuman preference · 2026-10-15 · rating

Claude Sonnet 3.7

1,284.37 rating

Confidence interval: 1,270.49 – 1,298.25

2196 votes

Category
polish
Evaluated configuration
claude-3-7-sonnet-20250219-thinking-32k
Evaluation type
human preference

1,278.61 rating

Confidence interval: 1,264.66 – 1,292.57

2177 votes

Category
polish
Evaluated configuration
claude-3-7-sonnet-20250219
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-15 · rating

Claude Sonnet 3.7

1,310.80 rating

Confidence interval: 1,300.46 – 1,321.14

3353 votes

Category
russian
Evaluated configuration
claude-3-7-sonnet-20250219-thinking-32k
Evaluation type
human preference

1,310.11 rating

Confidence interval: 1,300.08 – 1,320.13

3509 votes

Category
russian
Evaluated configuration
claude-3-7-sonnet-20250219
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-15 · rating

Claude Sonnet 3.7

1,298.32 rating

Confidence interval: 1,273.43 – 1,323.21

612 votes

Category
spanish
Evaluated configuration
claude-3-7-sonnet-20250219-thinking-32k
Evaluation type
human preference

1,289.66 rating

Confidence interval: 1,265.03 – 1,314.29

623 votes

Category
spanish
Evaluated configuration
claude-3-7-sonnet-20250219
Evaluation type
human preference

Run it locally

Inference · memory estimate

Claude Sonnet 3.7

Closed weights. No supported local deployment or GPU requirement is published.

API cost

USD / month

Standard text-token estimate. Excludes taxes, caching, batch discounts, reasoning and multimodal surcharges. Provider limits and regional availability may differ.

Claude Sonnet 3.7

Capabilities

Context window
200,000 tokens
Maximum output
64,000 tokens
Reasoning
Yes
Tool calling
Yes
Open weights
No
Knowledge cutoff
2024-10-31
Model license
Proprietary

Input modalities

  • text
  • image
  • pdf

Output modalities

  • text