A compact multimodal Gemma for images and short conversations. Supports over 140 languages and runs well on inexpensive GPUs.
| MMLU | 59.6 | |
|---|---|---|
| GSM8K | 38.4 | |
| HumanEval | 36 |
Which GPU does Gemma 3 4B need?
The entry configuration is 1x RTX3080TI, which delivers around 70 tokens per second. For more parallel requests and headroom on speed, go with 1x H200.
How much does Gemma 3 4B cost?
An hour of Gemma 3 4B runs from $0.28 to $8.95 depending on the GPU configuration. Billing is hourly: you pay for server time, not for tokens.
How do I call Gemma 3 4B from my code?
Once deployed you get an OpenAI-compatible endpoint: point any OpenAI client at the base_url you receive and pass gemma3:4b as the model name. Code written against the OpenAI API needs no changes.
Does the data stay private?
Yes. Gemma 3 4B runs on a dedicated GPU server inside a private environment: requests and responses never reach the model vendor and are not used for training.
What context window and license does Gemma 3 4B have?
Gemma 3 4B 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.