LIVE INTELLIGENCE · part of the Xentir AI Index

Claude Sonnet 4 (22 May 2025) vs Gemini 2.5 Flash (May 2025)

Name: Claude Sonnet 4 (22 May 2025) vs Gemini 2.5 Flash (May 2025) — benchmark results
Creator: Epoch AI
License: https://creativecommons.org/licenses/by/4.0/

Claude Sonnet 4 (22 May 2025) · Anthropic, released 2025-05-22
Gemini 2.5 Flash (May 2025) · Google DeepMind, released 2025-05-20

“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 2026-09-09 · Methodology · Data Sources

Claude Sonnet 4 (22 May 2025) leads on more of the 2 benchmarks where both are measured — Claude Sonnet 4 (22 May 2025) on 2. The remaining 7 of 9 tracked benchmarks cannot be compared: 4 have a result for only one of the two models, 3 for neither. Nothing on this page is estimated to fill those in.

Head-to-head

Measured on 2 of 9 tracked benchmarks for both models — Claude Sonnet 4 (22 May 2025) leads on 2.

Benchmark	Claude Sonnet 4 (22 May 2025)	Gemini 2.5 Flash (May 2025)	Leads
Epoch Capabilities Index (ECI)	141.7	141.6	Claude Sonnet 4 (22 May 2025)
SWE-bench Verified	Not measured	Not measured	Not measured on both
GPQA Diamond	0.783 (32k)	Not measured	Comparison unavailable — Gemini 2.5 Flash (May 2025) has no published result
MATH Level 5	0.844	Not measured	Comparison unavailable — Gemini 2.5 Flash (May 2025) has no published result
OTIS Mock AIME 2024-2025	0.711 (32k)	0.708	Claude Sonnet 4 (22 May 2025)
FrontierMath	0.041	Not measured	Comparison unavailable — Gemini 2.5 Flash (May 2025) has no published result
FrontierMath Tier 4	0.000	Not measured	Comparison unavailable — Gemini 2.5 Flash (May 2025) has no published result
SimpleQA Verified	Not measured	Not measured	Not measured on both
Chess Puzzles	Not measured	Not measured	Not measured on both

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.

What remains uncertain

Gemini 2.5 Flash (May 2025) has no published result on FrontierMath, FrontierMath Tier 4, GPQA Diamond, MATH Level 5.
3 tracked benchmarks have no published result for either model.

A missing result means Xentir has not found one published by the source it cites, not that the model performs badly. See Data Sources.

Reference price

Per 1M tokens	Claude Sonnet 4 (22 May 2025)	Gemini 2.5 Flash (May 2025)
Output	$15.00 stale	—

List price as published by the vendor. Xentir does not verify negotiated or enterprise rates, and price is not a measure of capability — see Data Sources.

Full profiles

Claude Sonnet 4 (22 May 2025) → · Gemini 2.5 Flash (May 2025) → · the full AI Model Rankings table