Multilingual and translation · Google

Gemma 3 4B

Hugging Face: google/gemma-3-4b-it · checked 2026-10-13

Small multimodal, multilingual Gemma.

Datasheet

Maker	Google
Parameters	4.30 BVerifiedsafetensors.total from the HF API Source
Licence	Gemma Terms of UseVerified
Access	Gated: accept the terms on Hugging Face to download
Task (HF pipeline)	image-text-to-text
First published on HF	2025-02-20
Last updated on HF	2025-03-21
Weights in memory	8.6 GB at 16-bit · 2.6 GB at ~4.85 bitsEstimateParameters × bits ÷ 8. Excludes context cache and runtime overhead; not every model has a 4-bit build.

Benchmark

1,273.2

LMArena text, non-English category (Elo-style) · #1 of 3 scored in this category

gemma-3-4b-it, non_english, 95% CI 1259.2–1287.1, 1,599 votes. Source

See the full multilingual and translation ranking

The description summarises the model card at https://huggingface.co/google/gemma-3-4b-it. Parameters and licence are read from the Hugging Face API; the weights-in-memory line is our arithmetic, not a measurement.