invokeable()

Gemini 3.7 Flash vs Grok 4.6

A custom comparison card. Numbers link back to their entries; data as of 2026-08-19. Build your own →

model
Gemini 3.7 Flash
Google
model
Grok 4.6
xAI
Context1M500K
Input $/M
Output $/M
Knowledge cutoff2026-032026-02
Available viaGemini API (AI Studio)xAI API
FrontierCode (score)43.6
DeepSWE65.3%
Terminal-Bench (accuracy)85.8
AutomationBench (accuracy)30.4
GDP.pdf (accuracy)34
GDM-MRCR (accuracy)97