SShogo Bench

Gemini 4 Argon vs GPT-5.6 Luna

Gemini 4 Argon leads GPT-5.6 Luna by 18.1 points on the Shogo Agent Index. Gemini 4 Argon 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

PillarGemini 4 ArgonGPT-5.6 Luna
Agentic coding7759
Long-horizon autonomy7861
Knowledge-work agents5939
Computer use6140
Security & context learning6151

Shared benchmarks

BenchmarkGemini 4 ArgonGPT-5.6 Luna
APEX-Agentsindex82.4%43.0%
Vending-Bench 2index$13,718$4,095
CursorBenchindex52.9%35.9%
ARC-AGI98.5%88.0%
ARC-AGI-291.3%59.5%
Chess Puzzles59.0%43.0%
CritPt27.1%20.6%
DTBench97.6%89.1%
FrontierCode47.5%39.8%
FrontierMath Tier 478.0%61.0%
FrontierMath Tiers 1-386.0%82.1%
GPQA Diamond96.4%91.6%
LMCA63.7%48.5%
MirrorCode58.9%21.1%
OTIS Mock AIME100.0%98.3%
ProofBench99.0%60.0%
SciCode61.8%53.6%
SimpleBench91.7%46.8%
SimpleQA Verified76.0%41.0%
WebDev Arena16801519
WeirdML v320.3%5.2%

Specs and price

Gemini 4 ArgonGPT-5.6 Luna
Agent Index69.5 (#6)51.4 (#34)
Agent $/M$0.84$0.10
Input / output $/M$2.00 / $10.00$0.20 / $1.20
Context1M1.1M
ReleasedOct 13, 2026Jul 9, 2026
WeightsClosedClosed
API hosts55

Don’t pick one. Route.

Shogo can run Gemini 4 Argon 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