# Google Ships Gemma 4 QAT Checkpoints — New Mobile Quantization Format Brings Gemma 4 E2B Under 1GB of RAM, Q4_0 Weights Land for E2B, E4B, 12B, and the 26B MoE on Hugging Face

> Gemma 4 shrinks to under 1GB on phones via a custom 2-bit mobile quantization format trained with QAT.

- Org: Google DeepMind
- Date: 2026-06-05
- Category: model, ecosystem
- Source: https://blog.google/innovation-and-ai/technology/developers-tools/quantization-aware-training-gemma-4/
- Canonical page: https://ai-tldr.dev/releases/google-gemma-4-qat-mobile/

## In short

Google releases quantization-aware-trained Gemma 4 weights in the popular Q4_0 format and a new mobile format that pushes E2B text-only under 1GB of RAM, with day-one llama.cpp, Ollama, MLX, vLLM, and SGLang support.

## What it is

A set of post-training checkpoints for Gemma 4 — Google's open-weight multimodal LLM family — that were compressed using Quantization-Aware Training (QAT) instead of standard post-training quantization. Two formats ship: the popular Q4_0 (4-bit) used by llama.cpp, and a new Google mobile format that uses 2-bit decoding layers, optimized KV caches, and static activations.

## How it works

QAT simulates the integer rounding of low-bit quantization during fine-tuning so the model is trained to be robust to the quality drop that post-training quantization (PTQ) normally introduces. Google reports QAT yields higher overall quality than its PTQ baselines at matching bit-widths. The mobile format pushes generation layers to 2 bits while keeping activations static, and pairs the encoder-free Gemma 4 architecture with a slimmer text-only path that drops the audio and vision encoders.

## Why it matters

The E2B text-only variant fits in under 1GB of RAM, putting a real multimodal-trained Google LLM inside the budget of mid-range phones and low-end laptops with no GPU. Day-one support across llama.cpp, Ollama, LM Studio, vLLM, SGLang, MLX, LiteRT-LM, Transformers.js, Hugging Face Transformers, and Unsloth means the new weights drop straight into the existing local-LLM tooling stack.

## Who it's for

Local-LLM hobbyists, on-device app builders, mobile ML engineers.

## Try it

huggingface.co/google/gemma-4-E2B-it-qat-mobile-transformers

## Links

- [Announcement](https://blog.google/innovation-and-ai/technology/developers-tools/quantization-aware-training-gemma-4/)
- [E2B Mobile Weights](https://huggingface.co/google/gemma-4-E2B-it-qat-mobile-transformers)
- [E4B Mobile Weights](https://huggingface.co/google/gemma-4-E4B-it-qat-mobile-transformers)
- [Q4_0 E2B Weights](https://huggingface.co/google/gemma-4-E2B-it-qat-q4_0-unquantized)
- [Gemma 4 12B Launch](https://blog.google/innovation-and-ai/technology/developers-tools/introducing-gemma-4-12b/)