LIVE INTELLIGENCE · part of the Xentir AI Index

Claude Haiku 4.5 vs Claude Opus 4

Name: Claude Haiku 4.5 vs Claude Opus 4 — benchmark results
Creator: Epoch AI
License: https://creativecommons.org/licenses/by/4.0/

Claude Haiku 4.5 · Anthropic, released 2025-10-15
Claude Opus 4 · Anthropic, released 2025-05-22

“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

Neither model leads on more of the 6 benchmarks where both are measured — Claude Haiku 4.5 on 3, Claude Opus 4 on 3. The remaining 3 of 9 tracked benchmarks cannot be compared: 3 have a result for only one of the two models, 0 for neither. Nothing on this page is estimated to fill those in.

Head-to-head

Measured on 6 of 9 tracked benchmarks for both models — Claude Haiku 4.5 leads on 3, Claude Opus 4 leads on 3.

Benchmark	Claude Haiku 4.5	Claude Opus 4	Leads
Epoch Capabilities Index (ECI)	142.4	142.7	Claude Opus 4
SWE-bench Verified	Not measured	0.707	Comparison unavailable — Claude Haiku 4.5 has no published result
GPQA Diamond	0.712 (32k)	0.763 (16k)	Claude Opus 4
MATH Level 5	0.964 (32k)	0.850	Claude Haiku 4.5
OTIS Mock AIME 2024-2025	0.667 (32k)	0.644 (27k)	Claude Haiku 4.5
FrontierMath	0.059 (32k)	0.045	Claude Haiku 4.5
FrontierMath Tier 4	0.021 (32k)	0.042 (27k)	Claude Opus 4
SimpleQA Verified	0.132	Not measured	Comparison unavailable — Claude Opus 4 has no published result
Chess Puzzles	0.080 (32k)	Not measured	Comparison unavailable — Claude Opus 4 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

Claude Haiku 4.5 has no published result on SWE-bench Verified.
Claude Opus 4 has no published result on Chess Puzzles, SimpleQA Verified.

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	Claude Opus 4
Output	$5.00 stale	$75.00 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 → · Claude Opus 4 → · the full AI Model Rankings table