ModelBeat
Models/GPT-5.5 Pro
All models
O

OpenAI: GPT-5.5 Pro

openai/gpt-5.5-pro

GPT-5.5 Pro is OpenAI’s high-capability model optimized for deep reasoning and accuracy on complex, high-stakes workloads.

Context
1.1M
Input
$30 / 1M
Output
$180 / 1M
Released
Apr 23, 2026
Modalities
fileimagetexttext

Benchmarks

Scores on standardized evaluations. Higher is better — and rank shows where GPT-5.5 Pro lands among all models tracked on Model Beat.

96.0
Intelligence Index
Epoch AI
96th percentile of tracked models

Reasoning

5 evals
ARC-AGI96.5%

Abstract visual reasoning

ARC-AGI-284.6%

Harder abstract reasoning

GPQA Diamond93.9%

Graduate-level scientific reasoning

SimpleBench76.9%

Common-sense trick questions

SimpleQA Verified64.5%

Factual accuracy & hallucination

Math

3 evals
AIME 2024/2025100.0%

Olympiad-qualifier math

FrontierMath52.4%

Research-level math problems

FrontierMath Tier 439.6%

Hardest research math

Model & benchmark data from Epoch AI (CC BY); pricing, specs & descriptions from OpenRouter.