Gemma 3n models

4 Gemma 3n models from Google on Hugging Face, from 5.4B to 8.4B parameters, published by Google in BF16, with a 32K-token context. The smallest official model, gemma-3n-E2B-it, needs about 12.2 GB of VRAM at its published precision; the cheapest live fit is RTX 5060 Ti at $0.173/hr.

Part of the Gemma series · Previous generation: Gemma 3 · Next generation: Gemma 4

Pick a size

One row per official Gemma 3n size: the VRAM it needs at each precision, the cheapest GPU that holds it today, and the KV cache for a 32K-token context.

ModelParametersNative VRAMFP8 VRAMINT4 VRAMLive GPU fit (native)Est. $/hrKV cache at 32K
gemma-3n-E2B-it5.4B12.2 GB6.1 GB3.0 GBRTX 5060 Ti$0.173/hr0.40 GB
gemma-3n-E4B-it7.8B17.5 GB8.8 GB4.4 GBRTX A5000$0.176/hr0.46 GB

VRAM is weights times a flat 1.2 overhead; the KV cache is a separate per-model figure at 16-bit, shown where the architecture is published. See the methodology.

Official models (2)

ModelParametersPublished asVRAM neededLive GPU fitEst. $/hr
gemma-3n-E2B-it5.4BBF1612.2 GBRTX 5060 Ti$0.173/hr
gemma-3n-E4B-it7.8BBF1617.5 GBRTX A5000$0.176/hr

Quantized versions (1)

ModelParametersPublished asVRAM neededLive GPU fitEst. $/hr
gemma-3n-E4B-it-MLX-bf167.8BBF1617.5 GBRTX A5000$0.176/hr

Fine-tunes and community models (1)

ModelParametersPublished asVRAM neededLive GPU fitEst. $/hr
gemma-3n-E4B-it8.4BBF1618.7 GBRTX A5000$0.176/hr

VRAM is for the precision the model is published in: the weights times a flat 1.2 overhead, with the KV cache not included. See the methodology. The fit shown is the lowest-priced single GPU type that holds the model at that precision, or the lowest-priced multi-GPU set (up to 8) when none does.

More from Google

Start ComfyUI on a rented GPU.

© 2026 Aquanode. All rights reserved.

All trademarks, logos and brand names are the property of their respective owners.