Deploy Google models on dedicated GPUs
19 Google models ready to deploy on GPU capacity reserved for you, from 268M to 31B parameters, covering chat & reasoning, embeddings, translation and safety & moderation. Private OpenAI-compatible endpoint, European GPUs, billed per GPU-hour.
| Model | Type | Params | Context | License |
|---|---|---|---|---|
| Chat & reasoning | 31B | 256K | Apache 2.0 | |
| Chat & reasoning | 8.0B | 128K | Apache 2.0 | |
| Chat & reasoning | 26B | 256K | Apache 2.0 | |
embeddinggemma-2trending Google | Embeddings | 744M | 256K | Apache 2.0 |
| Embeddings | 303M | — | Gemma Terms of Use | |
gemma-4-12B-itvision Google | Chat & reasoning | 12B | 256K | Apache 2.0 |
| Chat & reasoning | 1000M | — | Gemma Terms of Use | |
gemma-4-E2B-itvision Google | Chat & reasoning | 5.1B | 128K | Apache 2.0 |
gemma-3-4b-itvision Google | Chat & reasoning | 4.3B | 128K | Gemma Terms of Use |
medgemma-4b-itvision Google | Chat & reasoning | 4.3B | — | Custom license |
gemma-3-27b-itvision Google | Chat & reasoning | 27B | 128K | Gemma Terms of Use |
| Chat & reasoning | 268M | — | Gemma Terms of Use | |
gemma-3-12b-itvision Google | Chat & reasoning | 12B | 128K | Gemma Terms of Use |
medgemma-27b-itvision Google | Chat & reasoning | 27B | — | Custom license |
gemma-3n-E4B-itvision Google | Chat & reasoning | 7.8B | — | Gemma Terms of Use |
gemma-3n-E2B-itvision Google | Chat & reasoning | 5.4B | — | Gemma Terms of Use |
translategemma-12b-itvision Google | Translation | 12B | — | Gemma Terms of Use |
shieldgemma-2-4b-itvision Google | Safety & moderation | 4.3B | — | Gemma Terms of Use |
translategemma-27b-itvision Google | Translation | 27B | — | Gemma Terms of Use |
Specs from Hugging Face, synced 2026-10-27. Machine-readable version: https://sference.com/models/google.md
FAQ
Which Google models can I deploy on sference?
19 today, most requested first: gemma-4-31B-it, gemma-4-E4B-it, gemma-4-26B-A4B-it, embeddinggemma-2, embeddinggemma-300m, gemma-4-12B-it, gemma-3-1b-it and gemma-4-E2B-it, and 11 more. Any other Google checkpoint or a fine-tune of one can be requested with the form on this page.
How many GPUs do Google models need?
It depends on the model: the lineup runs from 268M to 31B parameters. Each model page lists the checkpoint size and the minimum footprint on 80 GB and 141 GB GPUs. You don't pick hardware on sference: you choose a performance tier and we size the deployment.
Can I use Google models commercially?
17 of the 19 listed models are under licenses that allow commercial use (Apache 2.0, Gemma Terms of Use and Custom license across the lineup). Each model page summarizes its license and links the full text.
How is a dedicated deployment priced?
Per GPU-hour on a monthly commitment, sized to your throughput and latency targets.
Request a Google model
Tell us which Google model and roughly how much traffic you expect. Fine-tunes of these models work the same way.
- 01We size it
A performance tier, the capacity behind it, and a quote for your volume.
- 02We benchmark it
Bring one workload with a sample of real traffic; we come back with latency and throughput numbers.
- 03It goes live
We provision the deployment. You call it with the same API and model id as before.