OpenAI logo

GPT-5.5

OpenAI · Released Apr 23, 2026

ProprietarycodingreasoningagentsCompare this model

$5

Input price / MTok

#25 of 27

$30

Output price / MTok

#25 of 27

1.0M

Context window

#8 of 28

89

Tokens / sec

#14 of 21

58.6

SWE-bench Pro

#5 of 9

93.6

GPQA Diamond

#3 of 19

78.7

OSWorld

#4 of 8

74

MRCR

#4 of 5

Summary

GPT-5.5 is a proprietary language model from OpenAI, released on Apr 23, 2026.

At $5 input and $30 output per million tokens, it ranks #25 of 27 for input price, placing it among the higher-priced models tracked here.

Its context window of 1.0M tokens ranks #8 of 28 for context depth in this dataset.

It targets coding, reasoning, agents workloads.

Throughput is measured at 89 tokens per second (ranked #14 of 21 for throughput).

OpenAI's current flagship; GPT-5.5 Instant became ChatGPT's default for free users May 5 2026.

Benchmarks

BenchmarkScoreMax%RankSource
SWE-bench Pro58.610058.6%#5/9OpenAI (self-reported in GPT-5.5 launch,...
GPQA Diamond93.610093.6%#3/19OpenAI (self-reported in GPT-5.5 launch,...
OSWorld78.710078.7%#4/8OpenAI (self-reported in GPT-5.5 launch,...
MRCR7410074.0%#4/5OpenAI (self-reported in GPT-5.5 launch,...

Similar Models

Compare all 6 models

Models with the most shared benchmark coverage and closest scores to GPT-5.5.

GPT-5.5
AnthropicClaude Sonnet 4.6
GoogleGemini 3.5 Flash
AnthropicClaude Opus 4.8
OpenAIGPT-5.4
TencentTencent Hy3
SWE-bench Pro#5 of 9
Professional software engineering tasks
GPQA Diamond#3 of 19
Graduate-level science reasoning
OSWorld#4 of 8
Computer-use and GUI automation
MRCR#4 of 5
Long-context retrieval and recall

Compare GPT-5.5 side by side

Pre-filled with 5 similar models

Open comparison

Speed source: https://artificialanalysis.ai/models/gpt-5-5