GeekRadar / The comparison desk

COMPARE.

AI models

openai

GPT-4 (0613)

Model evaluation

Capability & preference

Human preference · Arena

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

GPT-4 (0613) 1186.0 GPT-4 (0613) · 95% interval 1182.0–1190.0 · 88,723 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

GPT-4 (0613)

1,136.09 rating

Confidence interval: 1,125.69 – 1,146.49

8540 votes

Oct 28, 2026

Category: chinese
Evaluated configuration: GPT-4 (0613)
Evaluation type: human preference

Arena text preference / codinghuman preference · 2026-10-28 · rating

GPT-4 (0613)

1,187.11 rating

Confidence interval: 1,179.17 – 1,195.05

13719 votes

Oct 28, 2026

Category: coding
Evaluated configuration: GPT-4 (0613)
Evaluation type: human preference

Arena text preference / creative_writinghuman preference · 2026-10-28 · rating

GPT-4 (0613)

1,191.14 rating

Confidence interval: 1,182.87 – 1,199.41

13932 votes

Oct 28, 2026

Category: creative_writing
Evaluated configuration: GPT-4 (0613)
Evaluation type: human preference

Arena text preference / englishhuman preference · 2026-10-28 · rating

GPT-4 (0613)

1,205.27 rating

Confidence interval: 1,200.26 – 1,210.28

57478 votes

Oct 28, 2026

Category: english
Evaluated configuration: GPT-4 (0613)
Evaluation type: human preference

Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating

GPT-4 (0613)

1,058.26 rating

Confidence interval: 1,052.32 – 1,064.20

60653 votes

Oct 28, 2026

Category: exclude_ties
Evaluated configuration: GPT-4 (0613)
Evaluation type: human preference

Arena text preference / experthuman preference · 2026-10-28 · rating

GPT-4 (0613)

1,127.69 rating

Confidence interval: 1,114.95 – 1,140.42

3617 votes

Oct 28, 2026

Category: expert
Evaluated configuration: GPT-4 (0613)
Evaluation type: human preference

Arena text preference / frenchhuman preference · 2026-10-28 · rating

GPT-4 (0613)

1,170.43 rating

Confidence interval: 1,148.33 – 1,192.53

1545 votes

Oct 28, 2026

Category: french
Evaluated configuration: GPT-4 (0613)
Evaluation type: human preference

Arena text preference / germanhuman preference · 2026-10-28 · rating

GPT-4 (0613)

1,161.57 rating

Confidence interval: 1,145.35 – 1,177.78

2215 votes

Oct 28, 2026

Category: german
Evaluated configuration: GPT-4 (0613)
Evaluation type: human preference

Arena text preference / hard_promptshuman preference · 2026-10-28 · rating

GPT-4 (0613)

1,174.24 rating

Confidence interval: 1,167.82 – 1,180.65

23040 votes

Oct 28, 2026

Category: hard_prompts
Evaluated configuration: GPT-4 (0613)
Evaluation type: human preference

Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating

GPT-4 (0613)

1,188.12 rating

Confidence interval: 1,180.57 – 1,195.66

16244 votes

Oct 28, 2026

Category: hard_prompts_english
Evaluated configuration: GPT-4 (0613)
Evaluation type: human preference

Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating

GPT-4 (0613)

1,121.67 rating

Confidence interval: 1,112.35 – 1,130.99

8676 votes

Oct 28, 2026

Category: industry_business_and_management_and_financial_operations
Evaluated configuration: GPT-4 (0613)
Evaluation type: human preference

Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-28 · rating

GPT-4 (0613)

1,179.89 rating

Confidence interval: 1,171.84 – 1,187.94

13910 votes

Oct 28, 2026

Category: industry_entertainment_and_sports_and_media
Evaluated configuration: GPT-4 (0613)
Evaluation type: human preference

Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-28 · rating

GPT-4 (0613)

1,154.61 rating

Confidence interval: 1,142.32 – 1,166.90

4695 votes

Oct 28, 2026

Category: industry_legal_and_government
Evaluated configuration: GPT-4 (0613)
Evaluation type: human preference

Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-28 · rating

GPT-4 (0613)

1,153.33 rating

Confidence interval: 1,145.19 – 1,161.46

15252 votes

Oct 28, 2026

Category: industry_life_and_physical_and_social_science
Evaluated configuration: GPT-4 (0613)
Evaluation type: human preference

Arena text preference / industry_mathematicalhuman preference · 2026-10-28 · rating

GPT-4 (0613)

1,201.03 rating

Confidence interval: 1,192.22 – 1,209.84

9862 votes

Oct 28, 2026

Category: industry_mathematical
Evaluated configuration: GPT-4 (0613)
Evaluation type: human preference

Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-28 · rating

GPT-4 (0613)

1,123.19 rating

Confidence interval: 1,110.67 – 1,135.72

4768 votes

Oct 28, 2026

Category: industry_medicine_and_healthcare
Evaluated configuration: GPT-4 (0613)
Evaluation type: human preference

Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-28 · rating

GPT-4 (0613)

1,168.60 rating

Confidence interval: 1,161.95 – 1,175.25

23796 votes

Oct 28, 2026

Category: industry_software_and_it_services
Evaluated configuration: GPT-4 (0613)
Evaluation type: human preference

Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-28 · rating

GPT-4 (0613)

1,213.91 rating

Confidence interval: 1,207.07 – 1,220.74

22099 votes

Oct 28, 2026

Category: industry_writing_and_literature_and_language
Evaluated configuration: GPT-4 (0613)
Evaluation type: human preference

Arena text preference / instruction_followinghuman preference · 2026-10-28 · rating

GPT-4 (0613)

1,188.42 rating

Confidence interval: 1,182.44 – 1,194.40

29706 votes

Oct 28, 2026

Category: instruction_following
Evaluated configuration: GPT-4 (0613)
Evaluation type: human preference

Arena text preference / japanesehuman preference · 2026-10-28 · rating

GPT-4 (0613)

1,114.23 rating

Confidence interval: 1,094.25 – 1,134.22

1241 votes

Oct 28, 2026

Category: japanese
Evaluated configuration: GPT-4 (0613)
Evaluation type: human preference

Arena text preference / koreanhuman preference · 2026-10-28 · rating

GPT-4 (0613)

1,059.78 rating

Confidence interval: 1,037.42 – 1,082.15

1154 votes

Oct 28, 2026

Category: korean
Evaluated configuration: GPT-4 (0613)
Evaluation type: human preference

Arena text preference / longer_queryhuman preference · 2026-10-28 · rating

GPT-4 (0613)

1,187.56 rating

Confidence interval: 1,178.28 – 1,196.84

7523 votes

Oct 28, 2026

Category: longer_query
Evaluated configuration: GPT-4 (0613)
Evaluation type: human preference

Arena text preference / mathhuman preference · 2026-10-28 · rating

GPT-4 (0613)

1,216.96 rating

Confidence interval: 1,208.59 – 1,225.33

11181 votes

Oct 28, 2026

Category: math
Evaluated configuration: GPT-4 (0613)
Evaluation type: human preference

Arena text preference / multi_turnhuman preference · 2026-10-28 · rating

GPT-4 (0613)

1,184.68 rating

Confidence interval: 1,176.35 – 1,193.00

13168 votes

Oct 28, 2026

Category: multi_turn
Evaluated configuration: GPT-4 (0613)
Evaluation type: human preference

Arena text preference / non_englishhuman preference · 2026-10-28 · rating

GPT-4 (0613)

1,160.36 rating

Confidence interval: 1,154.56 – 1,166.16

31245 votes

Oct 28, 2026

Category: non_english
Evaluated configuration: GPT-4 (0613)
Evaluation type: human preference

Arena text preference / overallhuman preference · 2026-10-28 · rating

GPT-4 (0613)

1,185.98 rating

Confidence interval: 1,181.95 – 1,190.01

88723 votes

Oct 28, 2026

Category: overall
Evaluated configuration: GPT-4 (0613)
Evaluation type: human preference

Arena text preference / polishhuman preference · 2026-10-28 · rating

GPT-4 (0613)

1,219.37 rating

Confidence interval: 1,162.37 – 1,276.37

179 votes

Oct 28, 2026

Category: polish
Evaluated configuration: GPT-4 (0613)
Evaluation type: human preference

Arena text preference / russianhuman preference · 2026-10-28 · rating

GPT-4 (0613)

1,172.92 rating

Confidence interval: 1,162.86 – 1,182.98

5975 votes

Oct 28, 2026

Category: russian
Evaluated configuration: GPT-4 (0613)
Evaluation type: human preference

Arena text preference / spanishhuman preference · 2026-10-28 · rating

GPT-4 (0613)

1,167.90 rating

Confidence interval: 1,145.48 – 1,190.33

1364 votes

Oct 28, 2026

Category: spanish
Evaluated configuration: GPT-4 (0613)
Evaluation type: human preference

Run it locally

Inference · memory estimate

GPT-4 (0613)

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license: Proprietary