GeekRadar / The comparison desk

COMPARE.

AI models

openai

GPT-5.5 Instant

AI model Released May 5, 2026

Context window: 400,000 tokens
Maximum output: 128,000 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.

GPT-5.5 Instant 1419.2 GPT-5.5 Instant · 95% interval 1414.3–1424.1 · 27,128 votes · 2026-10-28

Model performance

Radar task score: four tests, equal weight. An experimental summary of these tasks, not a universal intelligence rating.

GPT-5.5 Instant

Tested configuration Radar task score

Unrated

2/4 core tests · not enough evidence

4 benchmark results for this configuration

Scientific reasoning

GPQA Diamond · 25%

Expert-written science questions; strict multiple-choice scoring. Chance accuracy is 25%.

GPT-5.5 Instant 82.5^*% Approx. 95% interval: 78.1–86.9% · 2026-08-02

Novel problem solving

Mystery Game Puzzles · 25%

Best-next-move puzzles in an undisclosed game variant. A narrow reasoning test, not general intelligence.

GPT-5.5 Instant Not available

Advanced mathematics

FrontierMath Tiers 1-3 · 25%

Private v2 mathematics problems with Python tools. Earlier versions and Tier 4 are excluded.

GPT-5.5 Instant 26.3^*% Approx. 95% interval: 21.2–31.4% · 2026-08-02

Factual answers

SimpleQA Verified · 25%

Correct factual answers under a guessing prompt, graded by a model. Not a hallucination or abstention score.

GPT-5.5 Instant Not available

Competition mathematics

OTIS Mock AIME 2024-2025 · Separate result

Competition problems with model-extracted answers checked against an exact key. Earlier grading procedures are excluded.

GPT-5.5 Instant 68.1^*% Approx. 95% interval: 55.9–80.2% · 2026-08-02

Research mathematics

FrontierMath Tier 4 v2 · Separate result

The hardest private FrontierMath problems, corrected v2 set, with Python tools. Separate from Tiers 1-3 in the core score.

GPT-5.5 Instant 2.4^*% Approx. 95% interval: 0.0–7.2% · 2026-08-02

Bars show reported task scores. Whiskers are approximate 95% intervals where standard errors are available. Different reasoning budgets are not equal-cost tests. Missing results are never zero.

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.

Test version

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

GPT-5.5 Instant

1,455.68 rating

Confidence interval: 1,440.65 – 1,470.71

1715 votes

Oct 28, 2026

Category: chinese
Evaluated configuration: GPT-5.5 Instant
Evaluation type: human preference

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

GPT-5.5 Instant

1,433.34 rating

Confidence interval: 1,425.75 – 1,440.93

7906 votes

Oct 28, 2026

Category: coding
Evaluated configuration: GPT-5.5 Instant
Evaluation type: human preference

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

GPT-5.5 Instant

1,417.65 rating

Confidence interval: 1,408.02 – 1,427.27

4341 votes

Oct 28, 2026

Category: creative_writing
Evaluated configuration: GPT-5.5 Instant
Evaluation type: human preference

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

GPT-5.5 Instant

1,417.65 rating

Confidence interval: 1,411.33 – 1,423.97

12916 votes

Oct 28, 2026

Category: english
Evaluated configuration: GPT-5.5 Instant
Evaluation type: human preference

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

GPT-5.5 Instant

1,408.26 rating

Confidence interval: 1,402.03 – 1,414.49

21282 votes

Oct 28, 2026

Category: exclude_ties
Evaluated configuration: GPT-5.5 Instant
Evaluation type: human preference

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

GPT-5.5 Instant

1,408.59 rating

Confidence interval: 1,396.60 – 1,420.57

2743 votes

Oct 28, 2026

Category: expert
Evaluated configuration: GPT-5.5 Instant
Evaluation type: human preference

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

GPT-5.5 Instant

1,427.21 rating

Confidence interval: 1,407.76 – 1,446.66

1039 votes

Oct 28, 2026

Category: french
Evaluated configuration: GPT-5.5 Instant
Evaluation type: human preference

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

GPT-5.5 Instant

1,410.51 rating

Confidence interval: 1,381.76 – 1,439.26

431 votes

Oct 28, 2026

Category: german
Evaluated configuration: GPT-5.5 Instant
Evaluation type: human preference

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

GPT-5.5 Instant

1,426.11 rating

Confidence interval: 1,420.30 – 1,431.92

18412 votes

Oct 28, 2026

Category: hard_prompts
Evaluated configuration: GPT-5.5 Instant
Evaluation type: human preference

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

GPT-5.5 Instant

1,423.05 rating

Confidence interval: 1,415.88 – 1,430.23

9176 votes

Oct 28, 2026

Category: hard_prompts_english
Evaluated configuration: GPT-5.5 Instant
Evaluation type: human preference

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

GPT-5.5 Instant

1,409.90 rating

Confidence interval: 1,401.35 – 1,418.46

5673 votes

Oct 28, 2026

Category: industry_business_and_management_and_financial_operations
Evaluated configuration: GPT-5.5 Instant
Evaluation type: human preference

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

GPT-5.5 Instant

1,405.89 rating

Confidence interval: 1,397.27 – 1,414.51

5804 votes

Oct 28, 2026

Category: industry_entertainment_and_sports_and_media
Evaluated configuration: GPT-5.5 Instant
Evaluation type: human preference

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

GPT-5.5 Instant

1,419.34 rating

Confidence interval: 1,406.36 – 1,432.32

2293 votes

Oct 28, 2026

Category: industry_legal_and_government
Evaluated configuration: GPT-5.5 Instant
Evaluation type: human preference

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

GPT-5.5 Instant

1,428.14 rating

Confidence interval: 1,418.75 – 1,437.53

4489 votes

Oct 28, 2026

Category: industry_life_and_physical_and_social_science
Evaluated configuration: GPT-5.5 Instant
Evaluation type: human preference

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

GPT-5.5 Instant

1,408.58 rating

Confidence interval: 1,393.25 – 1,423.90

1632 votes

Oct 28, 2026

Category: industry_mathematical
Evaluated configuration: GPT-5.5 Instant
Evaluation type: human preference

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

GPT-5.5 Instant

1,411.78 rating

Confidence interval: 1,398.12 – 1,425.45

2054 votes

Oct 28, 2026

Category: industry_medicine_and_healthcare
Evaluated configuration: GPT-5.5 Instant
Evaluation type: human preference

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

GPT-5.5 Instant

1,432.05 rating

Confidence interval: 1,425.34 – 1,438.76

11165 votes

Oct 28, 2026

Category: industry_software_and_it_services
Evaluated configuration: GPT-5.5 Instant
Evaluation type: human preference

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

GPT-5.5 Instant

1,421.02 rating

Confidence interval: 1,413.05 – 1,429.00

6677 votes

Oct 28, 2026

Category: industry_writing_and_literature_and_language
Evaluated configuration: GPT-5.5 Instant
Evaluation type: human preference

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

GPT-5.5 Instant

1,406.21 rating

Confidence interval: 1,399.21 – 1,413.21

9693 votes

Oct 28, 2026

Category: instruction_following
Evaluated configuration: GPT-5.5 Instant
Evaluation type: human preference

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

GPT-5.5 Instant

1,408.66 rating

Confidence interval: 1,371.29 – 1,446.03

267 votes

Oct 28, 2026

Category: japanese
Evaluated configuration: GPT-5.5 Instant
Evaluation type: human preference

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

GPT-5.5 Instant

1,392.48 rating

Confidence interval: 1,363.01 – 1,421.95

430 votes

Oct 28, 2026

Category: korean
Evaluated configuration: GPT-5.5 Instant
Evaluation type: human preference

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

GPT-5.5 Instant

1,421.91 rating

Confidence interval: 1,415.29 – 1,428.53

12541 votes

Oct 28, 2026

Category: longer_query
Evaluated configuration: GPT-5.5 Instant
Evaluation type: human preference

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

GPT-5.5 Instant

1,419.63 rating

Confidence interval: 1,404.13 – 1,435.13

1530 votes

Oct 28, 2026

Category: math
Evaluated configuration: GPT-5.5 Instant
Evaluation type: human preference

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

GPT-5.5 Instant

1,433.47 rating

Confidence interval: 1,424.48 – 1,442.47

5119 votes

Oct 28, 2026

Category: multi_turn
Evaluated configuration: GPT-5.5 Instant
Evaluation type: human preference

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

GPT-5.5 Instant

1,416.93 rating

Confidence interval: 1,410.77 – 1,423.08

14212 votes

Oct 28, 2026

Category: non_english
Evaluated configuration: GPT-5.5 Instant
Evaluation type: human preference

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

GPT-5.5 Instant

1,419.21 rating

Confidence interval: 1,414.28 – 1,424.14

27128 votes

Oct 28, 2026

Category: overall
Evaluated configuration: GPT-5.5 Instant
Evaluation type: human preference

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

GPT-5.5 Instant

1,413.99 rating

Confidence interval: 1,389.61 – 1,438.37

571 votes

Oct 28, 2026

Category: polish
Evaluated configuration: GPT-5.5 Instant
Evaluation type: human preference

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

GPT-5.5 Instant

1,430.21 rating

Confidence interval: 1,418.88 – 1,441.53

3024 votes

Oct 28, 2026

Category: russian
Evaluated configuration: GPT-5.5 Instant
Evaluation type: human preference

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

GPT-5.5 Instant

1,429.76 rating

Confidence interval: 1,408.10 – 1,451.42

813 votes

Oct 28, 2026

Category: spanish
Evaluated configuration: GPT-5.5 Instant
Evaluation type: human preference

Arena text preference / chinesehuman preference · 2026-10-15 · rating

GPT-5.5 Instant

1,456.29 rating

Confidence interval: 1,441.24 – 1,471.34

1712 votes

Oct 15, 2026

Category: chinese
Evaluated configuration: gpt-5.5-instant
Evaluation type: human preference

Arena text preference / codinghuman preference · 2026-10-15 · rating

GPT-5.5 Instant

1,433.13 rating

Confidence interval: 1,425.54 – 1,440.71

7925 votes

Oct 15, 2026

Category: coding
Evaluated configuration: gpt-5.5-instant
Evaluation type: human preference

Arena text preference / creative_writinghuman preference · 2026-10-15 · rating

GPT-5.5 Instant

1,417.73 rating

Confidence interval: 1,408.11 – 1,427.36

4350 votes

Oct 15, 2026

Category: creative_writing
Evaluated configuration: gpt-5.5-instant
Evaluation type: human preference

Arena text preference / englishhuman preference · 2026-10-15 · rating

GPT-5.5 Instant

1,417.60 rating

Confidence interval: 1,411.27 – 1,423.92

12927 votes

Oct 15, 2026

Category: english
Evaluated configuration: gpt-5.5-instant
Evaluation type: human preference

Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating

GPT-5.5 Instant

1,408.19 rating

Confidence interval: 1,401.96 – 1,414.42

21313 votes

Oct 15, 2026

Category: exclude_ties
Evaluated configuration: gpt-5.5-instant
Evaluation type: human preference

Arena text preference / experthuman preference · 2026-10-15 · rating

GPT-5.5 Instant

1,408.39 rating

Confidence interval: 1,396.41 – 1,420.36

2748 votes

Oct 15, 2026

Category: expert
Evaluated configuration: gpt-5.5-instant
Evaluation type: human preference

Arena text preference / frenchhuman preference · 2026-10-15 · rating

GPT-5.5 Instant

1,427.49 rating

Confidence interval: 1,408.12 – 1,446.87

1050 votes

Oct 15, 2026

Category: french
Evaluated configuration: gpt-5.5-instant
Evaluation type: human preference

Arena text preference / germanhuman preference · 2026-10-15 · rating

GPT-5.5 Instant

1,411.20 rating

Confidence interval: 1,382.39 – 1,440.00

430 votes

Oct 15, 2026

Category: german
Evaluated configuration: gpt-5.5-instant
Evaluation type: human preference

Arena text preference / hard_promptshuman preference · 2026-10-15 · rating

GPT-5.5 Instant

1,426.07 rating

Confidence interval: 1,420.26 – 1,431.88

18437 votes

Oct 15, 2026

Category: hard_prompts
Evaluated configuration: gpt-5.5-instant
Evaluation type: human preference

Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating

GPT-5.5 Instant

1,422.92 rating

Confidence interval: 1,415.75 – 1,430.09

9186 votes

Oct 15, 2026

Category: hard_prompts_english
Evaluated configuration: gpt-5.5-instant
Evaluation type: human preference

Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating

GPT-5.5 Instant

1,409.74 rating

Confidence interval: 1,401.18 – 1,418.30

5680 votes

Oct 15, 2026

Category: industry_business_and_management_and_financial_operations
Evaluated configuration: gpt-5.5-instant
Evaluation type: human preference

Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating

GPT-5.5 Instant

1,405.94 rating

Confidence interval: 1,397.32 – 1,414.56

5809 votes

Oct 15, 2026

Category: industry_entertainment_and_sports_and_media
Evaluated configuration: gpt-5.5-instant
Evaluation type: human preference

Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-15 · rating

GPT-5.5 Instant

1,419.47 rating

Confidence interval: 1,406.50 – 1,432.43

2301 votes

Oct 15, 2026

Category: industry_legal_and_government
Evaluated configuration: gpt-5.5-instant
Evaluation type: human preference

Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating

GPT-5.5 Instant

1,428.19 rating

Confidence interval: 1,418.80 – 1,437.59

4484 votes

Oct 15, 2026

Category: industry_life_and_physical_and_social_science
Evaluated configuration: gpt-5.5-instant
Evaluation type: human preference

Arena text preference / industry_mathematicalhuman preference · 2026-10-15 · rating

GPT-5.5 Instant

1,408.29 rating

Confidence interval: 1,392.95 – 1,423.64

1629 votes

Oct 15, 2026

Category: industry_mathematical
Evaluated configuration: gpt-5.5-instant
Evaluation type: human preference

Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating

GPT-5.5 Instant

1,411.96 rating

Confidence interval: 1,398.32 – 1,425.60

2065 votes

Oct 15, 2026

Category: industry_medicine_and_healthcare
Evaluated configuration: gpt-5.5-instant
Evaluation type: human preference

Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating

GPT-5.5 Instant

1,431.84 rating

Confidence interval: 1,425.13 – 1,438.55

11181 votes

Oct 15, 2026

Category: industry_software_and_it_services
Evaluated configuration: gpt-5.5-instant
Evaluation type: human preference

Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating

GPT-5.5 Instant

1,421.11 rating

Confidence interval: 1,413.13 – 1,429.09

6674 votes

Oct 15, 2026

Category: industry_writing_and_literature_and_language
Evaluated configuration: gpt-5.5-instant
Evaluation type: human preference

Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating

GPT-5.5 Instant

1,406.33 rating

Confidence interval: 1,399.33 – 1,413.33

9708 votes

Oct 15, 2026

Category: instruction_following
Evaluated configuration: gpt-5.5-instant
Evaluation type: human preference

Arena text preference / japanesehuman preference · 2026-10-15 · rating

GPT-5.5 Instant

1,409.53 rating

Confidence interval: 1,372.04 – 1,447.02

266 votes

Oct 15, 2026

Category: japanese
Evaluated configuration: gpt-5.5-instant
Evaluation type: human preference

Arena text preference / koreanhuman preference · 2026-10-15 · rating

GPT-5.5 Instant

1,392.39 rating

Confidence interval: 1,363.00 – 1,421.79

434 votes

Oct 15, 2026

Category: korean
Evaluated configuration: gpt-5.5-instant
Evaluation type: human preference

Arena text preference / longer_queryhuman preference · 2026-10-15 · rating

GPT-5.5 Instant

1,421.79 rating

Confidence interval: 1,415.18 – 1,428.41

12555 votes

Oct 15, 2026

Category: longer_query
Evaluated configuration: gpt-5.5-instant
Evaluation type: human preference

Arena text preference / mathhuman preference · 2026-10-15 · rating

GPT-5.5 Instant

1,419.37 rating

Confidence interval: 1,403.86 – 1,434.87

1530 votes

Oct 15, 2026

Category: math
Evaluated configuration: gpt-5.5-instant
Evaluation type: human preference

Arena text preference / multi_turnhuman preference · 2026-10-15 · rating

GPT-5.5 Instant

1,433.30 rating

Confidence interval: 1,424.31 – 1,442.29

5129 votes

Oct 15, 2026

Category: multi_turn
Evaluated configuration: gpt-5.5-instant
Evaluation type: human preference

Arena text preference / non_englishhuman preference · 2026-10-15 · rating

GPT-5.5 Instant

1,416.84 rating

Confidence interval: 1,410.69 – 1,423.00

14231 votes

Oct 15, 2026

Category: non_english
Evaluated configuration: gpt-5.5-instant
Evaluation type: human preference

Arena text preference / overallhuman preference · 2026-10-15 · rating

GPT-5.5 Instant

1,419.09 rating

Confidence interval: 1,414.16 – 1,424.02

27158 votes

Oct 15, 2026

Category: overall
Evaluated configuration: gpt-5.5-instant
Evaluation type: human preference

Arena text preference / polishhuman preference · 2026-10-15 · rating

GPT-5.5 Instant

1,414.77 rating

Confidence interval: 1,390.42 – 1,439.12

573 votes

Oct 15, 2026

Category: polish
Evaluated configuration: gpt-5.5-instant
Evaluation type: human preference

Arena text preference / russianhuman preference · 2026-10-15 · rating

GPT-5.5 Instant

1,429.85 rating

Confidence interval: 1,418.51 – 1,441.20

3023 votes

Oct 15, 2026

Category: russian
Evaluated configuration: gpt-5.5-instant
Evaluation type: human preference

Arena text preference / spanishhuman preference · 2026-10-15 · rating

GPT-5.5 Instant

1,429.68 rating

Confidence interval: 1,408.02 – 1,451.35

814 votes

Oct 15, 2026

Category: spanish
Evaluated configuration: gpt-5.5-instant
Evaluation type: human preference

Run it locally

Inference · memory estimate

GPT-5.5 Instant

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

API cost

USD / month

Input tokens / monthOutput tokens / month

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

GPT-5.5 Instant

Provider offer

Capabilities

Context window: 400,000 tokens
Maximum output: 128,000 tokens
Reasoning: Yes
Tool calling: Yes
Open weights: No
Knowledge cutoff: 2025-12-01
Model license: Proprietary

Input modalities

text
image
pdf

Output modalities

text

* Provisional data