NVIDIA · gemma-3

Cloud PlatformOpenAI compatible

NVIDIA NIM provides optimized inference for 190+ models on NVIDIA GPU infrastructure.

Active models
85+ 146 deprecated
Families
41
Headquarters🇺🇸 Santa Clara, CA
Founded1993
Latest release2026-10-20
API Base URL
Modelsbuild.nvidia.com/explore/discover
Capabilities
Model
gemma-3-12b-itgoogle/gemma-3-12b-it
gemma-3-4b-itgoogle/gemma-3-4b-it