Model reference · open weights

gemma-3-qat-q4_0

gemma-3-qat-q4_0 is an open-weight language model from Google. gemma-3-4b-it-qat-q4_0-gguf (GGUF) weighs 3.2 GB; the smallest configuration that runs it is RTX 3060 12 GB.

• Gemma 3 4B is an instruction-tuned multimodal model from Google that processes text and image inputs to generate text output.
• This specific release is a Quantization Aware Training (QAT) variant in GGUF format with Q4_0 quantization, designed to reduce memory requirements while preserving quality.
• The model features a 128K token context window, supports over 140 languages, and is distributed under the Gemma license.

Summary of the google/gemma-3-4b-it-qat-q4_0-gguf model card, 2026-10-01

Hugging Face ↗GitHub ↗Paper ↗