LIVE INTELLIGENCE · part of the Xentir AI Index

Claude Opus 4.7 vs GPT-5.4

Claude Opus 4.7 · Anthropic, released 2026-04-16
GPT-5.4 · OpenAI, released 2026-03-05

“Epoch AI, 'AI Benchmarking Hub'. Published online at epoch.ai. Retrieved from 'https://epoch.ai/benchmarks' [online resource].” · epoch.ai/benchmarks · Licensed CC BY 4.0 · data last verified 2026-07-26 · Methodology · Data Sources

Head-to-head

Measured on 8 of 9 tracked benchmarks for both models — Claude Opus 4.7 leads on 3, GPT-5.4 leads on 5.

Benchmark	Claude Opus 4.7	GPT-5.4	Leads
Epoch Capabilities Index (ECI)	156.1	156.1	GPT-5.4
SWE-bench Verified	0.835 (max)	0.769 (high)	Claude Opus 4.7
GPQA Diamond	0.902 (xhigh)	0.933 (xhigh)	GPT-5.4
MATH Level 5	Not measured	Not measured	Not measured on both
OTIS Mock AIME 2024-2025	0.978 (xhigh)	0.953 (xhigh)	Claude Opus 4.7
FrontierMath	0.438 (xhigh)	0.476 (xhigh)	GPT-5.4
FrontierMath Tier 4	0.229 (xhigh)	0.271 (xhigh)	GPT-5.4
SimpleQA Verified	0.506 (xhigh)	0.448 (xhigh)	Claude Opus 4.7
Chess Puzzles	0.300 (xhigh)	0.440 (xhigh)	GPT-5.4

Benchmark results measure specific tasks and should not be treated as a universal measure of model quality. A model with no row here for a benchmark has not been measured on it — that is left blank, never estimated.

Full profiles

Claude Opus 4.7 → · GPT-5.4 → · the full AI Model Rankings table