GeekRadar / The comparison desk

COMPARE.

AI models

openai

GPT-4 (0314)

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 (0314) 1206.0 GPT-4 (0314) · 95% interval 1201.3–1210.7 · 54,173 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 (0314)

1,183.87 rating

Confidence interval: 1,172.11 – 1,195.63

5849 votes

Oct 28, 2026

Category: chinese
Evaluated configuration: GPT-4 (0314)
Evaluation type: human preference

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

GPT-4 (0314)

1,208.40 rating

Confidence interval: 1,199.10 – 1,217.70

8306 votes

Oct 28, 2026

Category: coding
Evaluated configuration: GPT-4 (0314)
Evaluation type: human preference

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

GPT-4 (0314)

1,189.93 rating

Confidence interval: 1,180.07 – 1,199.79

8231 votes

Oct 28, 2026

Category: creative_writing
Evaluated configuration: GPT-4 (0314)
Evaluation type: human preference

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

GPT-4 (0314)

1,220.03 rating

Confidence interval: 1,214.28 – 1,225.79

35534 votes

Oct 28, 2026

Category: english
Evaluated configuration: GPT-4 (0314)
Evaluation type: human preference

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

GPT-4 (0314)

1,085.69 rating

Confidence interval: 1,078.69 – 1,092.70

37236 votes

Oct 28, 2026

Category: exclude_ties
Evaluated configuration: GPT-4 (0314)
Evaluation type: human preference

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

GPT-4 (0314)

1,148.61 rating

Confidence interval: 1,133.32 – 1,163.90

2160 votes

Oct 28, 2026

Category: expert
Evaluated configuration: GPT-4 (0314)
Evaluation type: human preference

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

GPT-4 (0314)

1,219.66 rating

Confidence interval: 1,193.67 – 1,245.65

839 votes

Oct 28, 2026

Category: french
Evaluated configuration: GPT-4 (0314)
Evaluation type: human preference

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

GPT-4 (0314)

1,198.24 rating

Confidence interval: 1,178.89 – 1,217.60

1265 votes

Oct 28, 2026

Category: german
Evaluated configuration: GPT-4 (0314)
Evaluation type: human preference

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

GPT-4 (0314)

1,199.97 rating

Confidence interval: 1,192.32 – 1,207.62

13935 votes

Oct 28, 2026

Category: hard_prompts
Evaluated configuration: GPT-4 (0314)
Evaluation type: human preference

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

GPT-4 (0314)

1,208.64 rating

Confidence interval: 1,199.86 – 1,217.42

9838 votes

Oct 28, 2026

Category: hard_prompts_english
Evaluated configuration: GPT-4 (0314)
Evaluation type: human preference

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

GPT-4 (0314)

1,145.05 rating

Confidence interval: 1,133.98 – 1,156.12

5128 votes

Oct 28, 2026

Category: industry_business_and_management_and_financial_operations
Evaluated configuration: GPT-4 (0314)
Evaluation type: human preference

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

GPT-4 (0314)

1,184.27 rating

Confidence interval: 1,174.66 – 1,193.88

8080 votes

Oct 28, 2026

Category: industry_entertainment_and_sports_and_media
Evaluated configuration: GPT-4 (0314)
Evaluation type: human preference

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

GPT-4 (0314)

1,191.57 rating

Confidence interval: 1,177.34 – 1,205.80

2897 votes

Oct 28, 2026

Category: industry_legal_and_government
Evaluated configuration: GPT-4 (0314)
Evaluation type: human preference

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

GPT-4 (0314)

1,176.82 rating

Confidence interval: 1,167.57 – 1,186.07

9610 votes

Oct 28, 2026

Category: industry_life_and_physical_and_social_science
Evaluated configuration: GPT-4 (0314)
Evaluation type: human preference

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

GPT-4 (0314)

1,227.61 rating

Confidence interval: 1,217.45 – 1,237.77

6112 votes

Oct 28, 2026

Category: industry_mathematical
Evaluated configuration: GPT-4 (0314)
Evaluation type: human preference

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

GPT-4 (0314)

1,145.07 rating

Confidence interval: 1,130.42 – 1,159.72

2913 votes

Oct 28, 2026

Category: industry_medicine_and_healthcare
Evaluated configuration: GPT-4 (0314)
Evaluation type: human preference

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

GPT-4 (0314)

1,194.60 rating

Confidence interval: 1,186.83 – 1,202.38

14549 votes

Oct 28, 2026

Category: industry_software_and_it_services
Evaluated configuration: GPT-4 (0314)
Evaluation type: human preference

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

GPT-4 (0314)

1,218.65 rating

Confidence interval: 1,210.35 – 1,226.94

12861 votes

Oct 28, 2026

Category: industry_writing_and_literature_and_language
Evaluated configuration: GPT-4 (0314)
Evaluation type: human preference

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

GPT-4 (0314)

1,200.88 rating

Confidence interval: 1,193.69 – 1,208.06

18087 votes

Oct 28, 2026

Category: instruction_following
Evaluated configuration: GPT-4 (0314)
Evaluation type: human preference

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

GPT-4 (0314)

1,136.62 rating

Confidence interval: 1,110.61 – 1,162.62

619 votes

Oct 28, 2026

Category: japanese
Evaluated configuration: GPT-4 (0314)
Evaluation type: human preference

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

GPT-4 (0314)

1,087.03 rating

Confidence interval: 1,054.89 – 1,119.16

429 votes

Oct 28, 2026

Category: korean
Evaluated configuration: GPT-4 (0314)
Evaluation type: human preference

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

GPT-4 (0314)

1,189.96 rating

Confidence interval: 1,178.45 – 1,201.47

4251 votes

Oct 28, 2026

Category: longer_query
Evaluated configuration: GPT-4 (0314)
Evaluation type: human preference

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

GPT-4 (0314)

1,230.06 rating

Confidence interval: 1,220.39 – 1,239.72

7052 votes

Oct 28, 2026

Category: math
Evaluated configuration: GPT-4 (0314)
Evaluation type: human preference

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

GPT-4 (0314)

1,205.91 rating

Confidence interval: 1,196.12 – 1,215.70

7992 votes

Oct 28, 2026

Category: multi_turn
Evaluated configuration: GPT-4 (0314)
Evaluation type: human preference

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

GPT-4 (0314)

1,194.14 rating

Confidence interval: 1,186.98 – 1,201.30

18639 votes

Oct 28, 2026

Category: non_english
Evaluated configuration: GPT-4 (0314)
Evaluation type: human preference

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

GPT-4 (0314)

1,205.98 rating

Confidence interval: 1,201.25 – 1,210.72

54173 votes

Oct 28, 2026

Category: overall
Evaluated configuration: GPT-4 (0314)
Evaluation type: human preference

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

GPT-4 (0314)

1,265.11 rating

Confidence interval: 1,201.33 – 1,328.88

119 votes

Oct 28, 2026

Category: polish
Evaluated configuration: GPT-4 (0314)
Evaluation type: human preference

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

GPT-4 (0314)

1,195.18 rating

Confidence interval: 1,183.01 – 1,207.36

3579 votes

Oct 28, 2026

Category: russian
Evaluated configuration: GPT-4 (0314)
Evaluation type: human preference

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

GPT-4 (0314)

1,196.65 rating

Confidence interval: 1,168.93 – 1,224.37

647 votes

Oct 28, 2026

Category: spanish
Evaluated configuration: GPT-4 (0314)
Evaluation type: human preference

Run it locally

Inference · memory estimate

GPT-4 (0314)

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license: Proprietary