Skip to content

Gemini 3.8 Flash vs Gemini 4 Argon

On the 33 use cases both models ran, Gemini 4 Argon won 33 to 0 (better outcome first, then lower cost).

Pass rate: Gemini 3.8 Flash 97% (33 runs) vs Gemini 4 Argon 100% (33 runs). Average cost per run: $0.37 vs $0.17.

MetricGemini 3.8 FlashGemini 4 Argon
Runs on shared use cases3333
Pass rate97% (32/33)100% (33/33)
Avg cost per run$0.37$0.17
Avg wall time4m3m
Step-ins per run0.180.21
Use cases won033

Use case by use case

Only use cases both models ran are compared; all settings of each model are grouped. How we score.