SShogo Bench

Gemini 3.8 Flash vs GPT-5.6 Luna

Gemini 3.8 Flash leads GPT-5.6 Luna by 4.8 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. Gemini 3.8 Flash wins 18 and GPT-5.6 Luna wins 4 of 22 shared benchmarks. GPT-5.6 Luna is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarGemini 3.8 FlashGPT-5.6 Luna
Agentic coding6359
Long-horizon autonomy6461
Knowledge-work agents4739
Computer use4440
Security & context learning5351

Shared benchmarks

BenchmarkGemini 3.8 FlashGPT-5.6 Luna
APEX-Agentsindex64.3%43.0%
DeepSWEindex73.8%67.2%
Vending-Bench 2index$5,094$4,095
CursorBenchindex39.6%35.9%
ARC-AGI98.5%88.0%
ARC-AGI-289.2%59.5%
Chess Puzzles58.0%43.0%
CritPt18.3%20.6%
DTBench95.7%89.1%
FrontierCode41.2%39.8%
FrontierMath Tier 422.0%61.0%
FrontierMath Tiers 1-368.4%82.1%
GPQA Diamond95.4%91.6%
LMCA52.9%48.5%
MirrorCode38.9%21.1%
OTIS Mock AIME98.9%98.3%
ProofBench48.0%60.0%
SciCode56.6%53.6%
SimpleBench82.4%46.8%
SimpleQA Verified69.7%41.0%
WebDev Arena15681519
WeirdML v37.4%5.2%

Specs and price

Gemini 3.8 FlashGPT-5.6 Luna
Agent Index56.2 (#21)51.4 (#34)
Agent $/M$0.35$0.10
Input / output $/M$0.75 / $3.75$0.20 / $1.20
Context1M1.1M
ReleasedSep 2, 2026Jul 9, 2026
WeightsClosedClosed
API hosts65

Don’t pick one. Route.

Shogo can run Gemini 3.8 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