QriticalBeta
Open Model Catalogue
Google
Google
Text generationVision

Gemma 4 12B

google/gemma-4-12B-it

Google multimodal · Apache 2.0

Deploy Gemma 4 12B nowView on HuggingFace
Parameters
12B
Min VRAM
≈ 28 GB
Hardware
1 × 80 GB GPU
OpenAI-compatible API with a per-workload key — drop-in for existing SDKs.
Dedicated GPUs — your model, your instance, no shared queues.
Billed by the second with auto-shutdown — never pay for idle time you didn't plan.