SShogo Bench

Claude Opus 5.5 vs GPT-5.6 Luna

Claude Opus 5.5 leads GPT-5.6 Luna by 19.5 points on the Shogo Agent Index. Claude Opus 5.5 wins 21 and GPT-5.6 Luna wins 0 of 21 shared benchmarks. GPT-5.6 Luna is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarClaude Opus 5.5GPT-5.6 Luna
Agentic coding8159
Long-horizon autonomy7661
Knowledge-work agents5839
Computer use6640
Security & context learning6251

Shared benchmarks

BenchmarkClaude Opus 5.5GPT-5.6 Luna
APEX-Agentsindex73.5%43.0%
Vending-Bench 2index$9,235$4,095
CursorBenchindex57.8%35.9%
ARC-AGI98.5%88.0%
ARC-AGI-293.3%59.5%
Chess Puzzles67.0%43.0%
CritPt31.7%20.6%
DTBench98.9%89.1%
FrontierCode54.6%39.8%
FrontierMath Tier 495.0%61.0%
FrontierMath Tiers 1-391.2%82.1%
GPQA Diamond94.1%91.6%
LMCA68.2%48.5%
MirrorCode95.6%21.1%
OTIS Mock AIME100.0%98.3%
ProofBench100.0%60.0%
SciCode66.9%53.6%
SimpleBench88.4%46.8%
SimpleQA Verified72.2%41.0%
WebDev Arena18201519
WeirdML v331.2%5.2%

Specs and price

Claude Opus 5.5GPT-5.6 Luna
Agent Index70.9 (#3)51.4 (#34)
Agent $/M$1.69$0.10
Input / output $/M$4.00 / $20.00$0.20 / $1.20
Context1M1.1M
ReleasedSep 22, 2026Jul 9, 2026
WeightsClosedClosed
API hosts65

Don’t pick one. Route.

Shogo can run Claude Opus 5.5 and GPT-5.6 Luna side by side and send each step of an agent to the one that’s cheaper for it.

Try both in Shogo