GeekRadar / The comparison desk
COMPARE.
AI models
openai

chatgpt-4o-latest-20250326

Model evaluation

Capability & preference

Human preference · Arena

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

chatgpt-4o-latest-20250326 1428.4 chatgpt-4o-latest-20250326 · 95% interval 1425.6–1431.2 · 81,914 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

chatgpt-4o-latest-20250326

1,456.58 rating

Confidence interval: 1,448.70 – 1,464.45

5874 votes

Category
chinese
Evaluated configuration
chatgpt-4o-latest-20250326
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-15 · rating

chatgpt-4o-latest-20250326

1,412.62 rating

Confidence interval: 1,407.58 – 1,417.67

15875 votes

Category
coding
Evaluated configuration
chatgpt-4o-latest-20250326
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating

chatgpt-4o-latest-20250326

1,405.23 rating

Confidence interval: 1,399.37 – 1,411.10

11530 votes

Category
creative_writing
Evaluated configuration
chatgpt-4o-latest-20250326
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating

chatgpt-4o-latest-20250326

1,434.45 rating

Confidence interval: 1,430.68 – 1,438.22

37268 votes

Category
english
Evaluated configuration
chatgpt-4o-latest-20250326
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating

chatgpt-4o-latest-20250326

1,421.15 rating

Confidence interval: 1,417.27 – 1,425.04

58399 votes

Category
exclude_ties
Evaluated configuration
chatgpt-4o-latest-20250326
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-15 · rating

chatgpt-4o-latest-20250326

1,400.48 rating

Confidence interval: 1,391.74 – 1,409.22

4599 votes

Category
expert
Evaluated configuration
chatgpt-4o-latest-20250326
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-15 · rating

chatgpt-4o-latest-20250326

1,446.49 rating

Confidence interval: 1,430.02 – 1,462.96

1403 votes

Category
french
Evaluated configuration
chatgpt-4o-latest-20250326
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-15 · rating

chatgpt-4o-latest-20250326

1,423.36 rating

Confidence interval: 1,408.91 – 1,437.80

1777 votes

Category
german
Evaluated configuration
chatgpt-4o-latest-20250326
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating

chatgpt-4o-latest-20250326

1,424.03 rating

Confidence interval: 1,420.36 – 1,427.70

38415 votes

Category
hard_prompts
Evaluated configuration
chatgpt-4o-latest-20250326
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating

chatgpt-4o-latest-20250326

1,428.22 rating

Confidence interval: 1,423.29 – 1,433.15

17292 votes

Category
hard_prompts_english
Evaluated configuration
chatgpt-4o-latest-20250326
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating

chatgpt-4o-latest-20250326

1,425.64 rating

Confidence interval: 1,420.10 – 1,431.19

13581 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
chatgpt-4o-latest-20250326
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating

chatgpt-4o-latest-20250326

1,408.81 rating

Confidence interval: 1,403.56 – 1,414.07

14848 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
chatgpt-4o-latest-20250326
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-15 · rating

chatgpt-4o-latest-20250326

1,444.55 rating

Confidence interval: 1,436.24 – 1,452.85

5308 votes

Category
industry_legal_and_government
Evaluated configuration
chatgpt-4o-latest-20250326
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating

chatgpt-4o-latest-20250326

1,442.05 rating

Confidence interval: 1,436.59 – 1,447.51

13719 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
chatgpt-4o-latest-20250326
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-15 · rating

chatgpt-4o-latest-20250326

1,406.18 rating

Confidence interval: 1,397.77 – 1,414.59

4893 votes

Category
industry_mathematical
Evaluated configuration
chatgpt-4o-latest-20250326
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating

chatgpt-4o-latest-20250326

1,453.92 rating

Confidence interval: 1,444.90 – 1,462.94

4608 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
chatgpt-4o-latest-20250326
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating

chatgpt-4o-latest-20250326

1,428.50 rating

Confidence interval: 1,424.29 – 1,432.70

26644 votes

Category
industry_software_and_it_services
Evaluated configuration
chatgpt-4o-latest-20250326
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating

chatgpt-4o-latest-20250326

1,414.67 rating

Confidence interval: 1,409.84 – 1,419.51

18581 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
chatgpt-4o-latest-20250326
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating

chatgpt-4o-latest-20250326

1,402.96 rating

Confidence interval: 1,398.58 – 1,407.33

22958 votes

Category
instruction_following
Evaluated configuration
chatgpt-4o-latest-20250326
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-15 · rating

chatgpt-4o-latest-20250326

1,405.13 rating

Confidence interval: 1,387.36 – 1,422.89

1286 votes

Category
japanese
Evaluated configuration
chatgpt-4o-latest-20250326
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-10-15 · rating

chatgpt-4o-latest-20250326

1,395.47 rating

Confidence interval: 1,379.24 – 1,411.70

1387 votes

Category
korean
Evaluated configuration
chatgpt-4o-latest-20250326
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-15 · rating

chatgpt-4o-latest-20250326

1,412.27 rating

Confidence interval: 1,407.41 – 1,417.14

17779 votes

Category
longer_query
Evaluated configuration
chatgpt-4o-latest-20250326
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-15 · rating

chatgpt-4o-latest-20250326

1,406.95 rating

Confidence interval: 1,399.27 – 1,414.63

5784 votes

Category
math
Evaluated configuration
chatgpt-4o-latest-20250326
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-15 · rating

chatgpt-4o-latest-20250326

1,454.21 rating

Confidence interval: 1,448.76 – 1,459.67

14283 votes

Category
multi_turn
Evaluated configuration
chatgpt-4o-latest-20250326
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating

chatgpt-4o-latest-20250326

1,418.84 rating

Confidence interval: 1,415.34 – 1,422.33

44590 votes

Category
non_english
Evaluated configuration
chatgpt-4o-latest-20250326
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating

chatgpt-4o-latest-20250326

1,428.43 rating

Confidence interval: 1,425.65 – 1,431.22

81914 votes

Category
overall
Evaluated configuration
chatgpt-4o-latest-20250326
Evaluation type
human preference
Arena text preference / polishhuman preference · 2026-10-15 · rating

chatgpt-4o-latest-20250326

1,447.61 rating

Confidence interval: 1,437.10 – 1,458.11

3623 votes

Category
polish
Evaluated configuration
chatgpt-4o-latest-20250326
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-15 · rating

chatgpt-4o-latest-20250326

1,429.15 rating

Confidence interval: 1,422.46 – 1,435.84

7796 votes

Category
russian
Evaluated configuration
chatgpt-4o-latest-20250326
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-15 · rating

chatgpt-4o-latest-20250326

1,433.66 rating

Confidence interval: 1,419.78 – 1,447.55

1926 votes

Category
spanish
Evaluated configuration
chatgpt-4o-latest-20250326
Evaluation type
human preference

Run it locally

Inference · memory estimate

chatgpt-4o-latest-20250326

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license
Proprietary