SShogo Bench

Claude Sonnet 5 vs GPT-5.6 Luna

Claude Sonnet 5 leads GPT-5.6 Luna by 1.0 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. Claude Sonnet 5 wins 11 and GPT-5.6 Luna wins 8 of 19 shared benchmarks. GPT-5.6 Luna is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarClaude Sonnet 5GPT-5.6 Luna
Agentic coding5759
Long-horizon autonomy6561
Knowledge-work agents4339
Computer use4040
Security & context learning5151

Shared benchmarks

BenchmarkClaude Sonnet 5GPT-5.6 Luna
APEX-Agentsindex54.5%43.0%
DeepSWEindex53.8%67.2%
Vending-Bench 2index$6,378$4,095
CursorBenchindex34.1%35.9%
Chess Puzzles45.0%43.0%
CritPt16.9%20.6%
DTBench92.5%89.1%
FrontierCode42.7%39.8%
FrontierMath Tier 429.3%61.0%
FrontierMath Tiers 1-365.6%82.1%
GPQA Diamond90.5%91.6%
LMCA50.0%48.5%
MirrorCode63.3%21.1%
OTIS Mock AIME94.7%98.3%
ProofBench77.0%60.0%
SciCode54.3%53.6%
SimpleBench60.6%46.8%
SimpleQA Verified33.7%41.0%
WebDev Arena15371519

Specs and price

Claude Sonnet 5GPT-5.6 Luna
Agent Index52.4 (#32)51.4 (#34)
Agent $/M$0.92$0.10
Input / output $/M$2.00 / $10.00$0.20 / $1.20
Context1M1.1M
ReleasedJun 30, 2026Jul 9, 2026
WeightsClosedClosed
API hosts55

Don’t pick one. Route.

Shogo can run Claude Sonnet 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