SShogo Bench

Gemini 3.8 Flash vs GPT-5.5

GPT-5.5 leads Gemini 3.8 Flash by 0.5 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. Gemini 3.8 Flash wins 10 and GPT-5.5 wins 11 of 21 shared benchmarks. Gemini 3.8 Flash is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarGemini 3.8 FlashGPT-5.5
Agentic coding6363
Long-horizon autonomy6467
Knowledge-work agents4746
Computer use4443
Security & context learning5355

Shared benchmarks

BenchmarkGemini 3.8 FlashGPT-5.5
APEX-Agentsindex64.3%55.1%
DeepSWEindex73.8%67.0%
Vending-Bench 2index$5,094$7,524
Remote Labor Indexindex5.8%6.3%
ARC-AGI98.5%95.0%
ARC-AGI-289.2%85.0%
Chess Puzzles58.0%63.0%
CritPt18.3%27.1%
DTBench95.7%96.0%
FrontierCode41.2%43.0%
FrontierMath Tier 422.0%72.5%
FrontierMath Tiers 1-368.4%85.3%
GPQA Diamond95.4%94.0%
LMCA52.9%54.3%
MirrorCode38.9%23.3%
OTIS Mock AIME98.9%100.0%
ProofBench48.0%50.0%
SciCode56.6%56.1%
SimpleBench82.4%69.0%
SimpleQA Verified69.7%63.0%
WebDev Arena15681510

Specs and price

Gemini 3.8 FlashGPT-5.5
Agent Index56.2 (#21)56.7 (#19)
Agent $/M$0.35$2.55
Input / output $/M$0.75 / $3.75$5.00 / $30.00
Context1M1.1M
ReleasedSep 2, 2026Apr 23, 2026
WeightsClosedClosed
API hosts64

Don’t pick one. Route.

Shogo can run Gemini 3.8 Flash and GPT-5.5 side by side and send each step of an agent to the one that’s cheaper for it.

Try both in Shogo