DataLearner logo

GPT-5.5 ProvsGPT-5.4 Pro

Across 5 shared benchmarks, GPT-5.5 Pro leads overall: GPT-5.5 Pro wins 4, GPT-5.4 Pro wins 1, with 0 ties and an average score difference of +5.34.

OpenAI
GPT-5.5 Pro

OpenAI · 2026-04-23 · Reasoning model

OpenAI
GPT-5.4 Pro

OpenAI · 2026-03-05 · Multimodal model

GPT-5.5 Pro4 wins(80%)(20%)1 winGPT-5.4 Pro

Benchmark scores

Grouped by capability, sorted by largest gap within each. 5 shared benchmarks.

General Knowledge

GPT-5.5 Pro 2/2
BenchmarkGPT-5.5 ProGPT-5.4 ProDiff
ARC-AGI-196.5015 / 147Thinking High (No Tools)94.5026 / 147Thinking High (No Tools)+2
CritPt30.605 / 200Extra-High (No Tools)307 / 200Extra-High (No Tools)+0.60

Math and Reasoning

GPT-5.5 Pro 2/2
BenchmarkGPT-5.5 ProGPT-5.4 ProDiff
FrontierMath Tier 4 v278.055 / 42Extra-High (No Tools)58.5410 / 42Extra-High (No Tools)+19.51
FrontierMath v287.722 / 58Extra-High (No Tools)82.467 / 58Extra-High (No Tools)+5.26

General Evaluation

GPT-5.4 Pro 1/1
BenchmarkGPT-5.5 ProGPT-5.4 ProDiff
GPQA Diamond93.9218 / 462Extra-High (No Tools)94.606 / 462Extra-High (No Tools)-0.68

Specs

FieldGPT-5.5 ProGPT-5.4 Pro
PublisherOpenAIOpenAI
Release date2026-04-232026-03-05
Model typeReasoning modelMultimodal model
ArchitectureDenseDense
ParametersNot availableNot available
Context length1000K1M
Max output128K125K

API pricing

Prices use DataLearner records when available; missing fields are not inferred.

ItemGPT-5.5 ProGPT-5.4 Pro
Text input$30 / 1M tokens$30 / 1M tokens
Text output$180 / 1M tokens$180 / 1M tokens

Summary

  • GPT-5.5 Proleads in:General Knowledge (2/2), Math and Reasoning (2/2)
  • GPT-5.4 Proleads in:General Evaluation (1/1)

On average across the 5 shared benchmarks, GPT-5.5 Pro scores 5.34 higher.

Largest single-benchmark gap: FrontierMath Tier 4 v2 — GPT-5.5 Pro 78.05 vs GPT-5.4 Pro 58.54 (+19.51).

Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.