SShogo Bench

Gemini 3.8 Flash vs Grok 4.7

Gemini 3.8 Flash leads Grok 4.7 by 1.4 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. Gemini 3.8 Flash wins 12 and Grok 4.7 wins 10 of 22 shared benchmarks. Gemini 3.8 Flash is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarGemini 3.8 FlashGrok 4.7
Agentic coding6358
Long-horizon autonomy6469
Knowledge-work agents4747
Computer use4439
Security & context learning5350

Shared benchmarks

BenchmarkGemini 3.8 FlashGrok 4.7
APEX-Agentsindex64.3%70.0%
Vending-Bench 2index$5,094$10,537
FrontierSWEindex19.6%29.5%
CursorBenchindex39.6%46.3%
ARC-AGI98.5%90.2%
ARC-AGI-289.2%61.4%
Chess Puzzles58.0%41.0%
CritPt18.3%18.0%
DTBench95.7%96.0%
FrontierCode41.2%47.6%
FrontierMath Tier 422.0%17.1%
FrontierMath Tiers 1-368.4%53.0%
GeoBench86.0%61.0%
GPQA Diamond95.4%92.7%
LMCA52.9%49.4%
MirrorCode38.9%54.4%
OTIS Mock AIME98.9%98.1%
ProofBench48.0%26.0%
SciCode56.6%57.8%
SimpleQA Verified69.7%56.0%
WebDev Arena15681638
WeirdML v37.4%8.5%

Specs and price

Gemini 3.8 FlashGrok 4.7
Agent Index56.2 (#21)54.8 (#27)
Agent $/M$0.35$0.98
Input / output $/M$0.75 / $3.75$2.00 / $6.00
Context1M500K
ReleasedSep 2, 2026Sep 21, 2026
WeightsClosedClosed
API hosts65

Don’t pick one. Route.

Shogo can run Gemini 3.8 Flash and Grok 4.7 side by side and send each step of an agent to the one that’s cheaper for it.

Try both in Shogo