supermarket

Vision-language

GPT-5.4 Pro

By OpenAI, in the GPT-5 family.

Released
5 Mar 2026
Status
Legacy
Context
1.1M
Max output
128K
Input per 1M tokens
$30.00
Output per 1M tokens
$180.00
Open weights
No
License
proprietary
Parameters
—
Knowledge cutoff
31 Aug 2025
Reasoning
Yes
Modalities
text, image → text

Category scores

Out of 100, from independent and third-party results only, by method category-2026-09-29.1.

Category scores
CategoryScoreRankCoverage
Knowledge & reasoning62.752 of 4
Factuality & hallucination46.3301 of 1

Benchmark results

Every result we hold for this model, with who ran it and where it was published.

Knowledge & reasoning

Knowledge & reasoning results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
Chess PuzzlesPuzzles solved58.6%Independentxhigh19 Mar 2026Epoch AI
SimpleBenchScore (AVG@5)74.1%Third partyunknown—SimpleBench Leaderboard via Epoch AI

Scientific reasoning

Scientific reasoning results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
GPQA DiamondAccuracy94.6%Independentxhigh20 Mar 2026Epoch AI

Mathematics

Mathematics results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
FrontierMath Tier 4Accuracy58.5%Independentxhigh13 Jun 2026Epoch AI
FrontierMath Tiers 1-3Accuracy82.5%Independentxhigh13 Jun 2026Epoch AI
FrontierMath Tiers 1-3 (v1)Accuracy50.0%Independentxhigh6 Mar 2026Epoch AI

Coding & software engineering

Coding & software engineering results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
WeirdML (v2)Average accuracy57.4%Third partynone—https://htihle.github.io/weirdml.html via Epoch AI

Factuality & hallucination

Factuality & hallucination results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
SimpleQA VerifiedAccuracy (% correct)46.3%Independentxhigh27 Aug 2026Epoch AI

Other GPT-5 versions

Sources