LIVE INTELLIGENCE · part of the Xentir AI Index

Claude Opus 4.6 vs GPT-5.6 Luna

Name: Claude Opus 4.6 vs GPT-5.6 Luna — benchmark results
Creator: Epoch AI
License: https://creativecommons.org/licenses/by/4.0/

Updated 10 September 2026

What remains uncertain

GPT-5.6 Luna has no published result on FrontierMath, FrontierMath Tier 4, SWE-bench Verified.

Claude Opus 4.6 · Anthropic, released 2026-02-05
GPT-5.6 Luna · OpenAI, released 2026-07-09

“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 2026-09-10 · Methodology · Data Sources

GPT-5.6 Luna leads on more of the 5 benchmarks where both are measured — Claude Opus 4.6 on 1, GPT-5.6 Luna on 4. The remaining 4 of 9 tracked benchmarks cannot be compared: 3 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 5 of 9 tracked benchmarks for both models — Claude Opus 4.6 leads on 1, GPT-5.6 Luna leads on 4.

Benchmark	Claude Opus 4.6	GPT-5.6 Luna	Leads
Epoch Capabilities Index (ECI)	155.3	156.3	GPT-5.6 Luna
SWE-bench Verified	0.772	Not measured	Comparison unavailable — GPT-5.6 Luna has no published result
GPQA Diamond	0.905 (32k)	0.916 (max)	GPT-5.6 Luna
MATH Level 5	Not measured	Not measured	Not measured on both
OTIS Mock AIME 2024-2025	0.944 (64k)	0.983 (max)	GPT-5.6 Luna
FrontierMath	0.407 (max)	Not measured	Comparison unavailable — GPT-5.6 Luna has no published result
FrontierMath Tier 4	0.229 (max)	Not measured	Comparison unavailable — GPT-5.6 Luna has no published result
SimpleQA Verified	0.470 (max)	0.410 (max)	Claude Opus 4.6
Chess Puzzles	0.170 (32k)	0.400 (max)	GPT-5.6 Luna

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

GPT-5.6 Luna has no published result on FrontierMath, FrontierMath Tier 4, SWE-bench 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 Opus 4.6	GPT-5.6 Luna
Output	$25.00 stale	$1.20

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 Opus 4.6 → · GPT-5.6 Luna → · the full AI Model Rankings table