SShogo Bench

Gemini 3.6 Flash vs GPT-5.6 Luna

GPT-5.6 Luna leads Gemini 3.6 Flash by 3.1 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. Gemini 3.6 Flash wins 8 and GPT-5.6 Luna wins 10 of 18 shared benchmarks. GPT-5.6 Luna is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarGemini 3.6 FlashGPT-5.6 Luna
Agentic coding5359
Long-horizon autonomy5861
Knowledge-work agents3939
Computer use3740
Security & context learning4951

Shared benchmarks

BenchmarkGemini 3.6 FlashGPT-5.6 Luna
APEX-Agentsindex46.9%43.0%
DeepSWEindex46.7%67.2%
ARC-AGI91.2%88.0%
ARC-AGI-260.4%59.5%
Chess Puzzles42.0%43.0%
CritPt10.6%20.6%
DTBench95.5%89.1%
FrontierCode34.4%39.8%
FrontierMath Tier 422.0%61.0%
FrontierMath Tiers 1-358.9%82.1%
GPQA Diamond94.1%91.6%
LMCA44.9%48.5%
MirrorCode23.3%21.1%
OTIS Mock AIME94.2%98.3%
ProofBench36.0%60.0%
SciCode53.4%53.6%
SimpleQA Verified66.2%41.0%
WebDev Arena15371519

Specs and price

Gemini 3.6 FlashGPT-5.6 Luna
Agent Index48.3 (#39)51.4 (#34)
Agent $/M$0.35$0.10
Input / output $/M$0.75 / $3.75$0.20 / $1.20
Context1M1.1M
ReleasedJul 21, 2026Jul 9, 2026
WeightsClosedClosed
API hosts65

Don’t pick one. Route.

Shogo can run Gemini 3.6 Flash 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