LIVE INTELLIGENCE · part of the Xentir AI Index

Claude Haiku 4.5 vs Gemini 2.5 Pro (May 2025)

Name: Claude Haiku 4.5 vs Gemini 2.5 Pro (May 2025) — benchmark results
Creator: Epoch AI
License: https://creativecommons.org/licenses/by/4.0/

Claude Haiku 4.5 · Anthropic, released 2025-10-15
Gemini 2.5 Pro (May 2025) · Google DeepMind, released 2025-05-06

“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 Haiku 4.5 leads on more of the 3 benchmarks where both are measured — Claude Haiku 4.5 on 2, Gemini 2.5 Pro (May 2025) on 1. The remaining 6 of 9 tracked benchmarks cannot be compared: 5 have a result for only one of the two models, 1 for neither. Nothing on this page is estimated to fill those in.

Head-to-head

Measured on 3 of 9 tracked benchmarks for both models — Claude Haiku 4.5 leads on 2, Gemini 2.5 Pro (May 2025) leads on 1.

Benchmark	Claude Haiku 4.5	Gemini 2.5 Pro (May 2025)	Leads
Epoch Capabilities Index (ECI)	142.4	142.5	Gemini 2.5 Pro (May 2025)
SWE-bench Verified	Not measured	Not measured	Not measured on both
GPQA Diamond	0.712 (32k)	0.667	Claude Haiku 4.5
MATH Level 5	0.964 (32k)	0.959	Claude Haiku 4.5
OTIS Mock AIME 2024-2025	0.667 (32k)	Not measured	Comparison unavailable — Gemini 2.5 Pro (May 2025) has no published result
FrontierMath	0.059 (32k)	Not measured	Comparison unavailable — Gemini 2.5 Pro (May 2025) has no published result
FrontierMath Tier 4	0.021 (32k)	Not measured	Comparison unavailable — Gemini 2.5 Pro (May 2025) has no published result
SimpleQA Verified	0.132	Not measured	Comparison unavailable — Gemini 2.5 Pro (May 2025) has no published result
Chess Puzzles	0.080 (32k)	Not measured	Comparison unavailable — Gemini 2.5 Pro (May 2025) has no published result

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 Pro (May 2025) has no published result on Chess Puzzles, FrontierMath, FrontierMath Tier 4, OTIS Mock AIME 2024-2025, SimpleQA Verified.
1 tracked benchmark has 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 Haiku 4.5	Gemini 2.5 Pro (May 2025)
Output	$5.00 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 Haiku 4.5 → · Gemini 2.5 Pro (May 2025) → · the full AI Model Rankings table