A balanced multimodal Gemma for documents, dialogue and visual analysis. Stronger than compact variants while remaining single-GPU.
| MMLU | 74.5 | |
|---|---|---|
| GSM8K | 71 | |
| HumanEval | 45.7 |
Which GPU does Gemma 3 12B need?
The entry configuration is 1x RTX3090, which delivers around 36 tokens per second. For more parallel requests and headroom on speed, go with 1x H200.
How much does Gemma 3 12B cost?
An hour of Gemma 3 12B runs from $0.45 to $8.26 depending on the GPU configuration. Billing is hourly: you pay for server time, not for tokens.
How do I call Gemma 3 12B from my code?
Once deployed you get an OpenAI-compatible endpoint: point any OpenAI client at the base_url you receive and pass gemma3:12b as the model name. Code written against the OpenAI API needs no changes.
Does the data stay private?
Yes. Gemma 3 12B runs on a dedicated GPU server inside a private loop: requests and responses never reach the model vendor and are not used for training.
What context window and license does Gemma 3 12B have?
Gemma 3 12B holds 32K tokens of context per request and ships under the Gemma Terms of Use license, so you can use it commercially on your own hosting terms.