LIVE INTELLIGENCE · part of the Xentir AI Index

Claude Sonnet 4.5 vs GPT-5.4 nano

Claude Sonnet 4.5 · Anthropic, released 2025-09-29
GPT-5.4 nano · OpenAI, released 2026-03-17

“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 7 of 9 tracked benchmarks for both models — Claude Sonnet 4.5 leads on 3, GPT-5.4 nano leads on 4.

Benchmark	Claude Sonnet 4.5	GPT-5.4 nano	Leads
Epoch Capabilities Index (ECI)	146.8	146.6	Claude Sonnet 4.5
SWE-bench Verified	0.713	Not measured	Not measured on both
GPQA Diamond	0.823 (59k)	0.785 (high)	Claude Sonnet 4.5
MATH Level 5	0.977 (32k)	Not measured	Not measured on both
OTIS Mock AIME 2024-2025	0.778 (59k)	0.878 (high)	GPT-5.4 nano
FrontierMath	0.152 (32k)	0.259 (high)	GPT-5.4 nano
FrontierMath Tier 4	0.042 (32k)	0.062 (high)	GPT-5.4 nano
SimpleQA Verified	0.236 (59k)	0.120 (high)	Claude Sonnet 4.5
Chess Puzzles	0.120 (32k)	0.300 (high)	GPT-5.4 nano

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 Sonnet 4.5 → · GPT-5.4 nano → · the full AI Model Rankings table