Opening article
Welcome Gemma 3: Google's all new multimodal, multilingual, long context open LLM. Today Google releases Gemma 3 , a new iteration of their Gemma family of models. The models range from 1B to 27B parameters, have a context window up to 128k tokens, can accept images and text, and support 140+ languages. As is the case with other VLMs (vision-language models), Gemma 3 generates text in response to the user inputs, which may consist of text and, optionally, images. Gemma 3 models use SigLIP as an image encoder, which encodes images into tokens that are ingested into the language model.
Today Google releases Gemma 3 , a new iteration of their Gemma family of models. The models range from 1B to 27B parameters, have a context window up to 128k tokens, can accept images and text, and support 140+ languages.
All the models are on the Hub and tightly integrated with the Hugging Face ecosystem.
Both pre-trained and instruction tuned models are released. Gemma-3-4B-IT beats Gemma-2-27B IT, while Gemma-3-27B-IT beats Gemini 1.5-Pro across benchmarks .
Gemma 3 is Google's latest iteration of open weight LLMs. It comes in four sizes, 1 billion , 4 billion , 12 billion , and 27 billion parameters with base (pre-trained) and instruction-tuned versions. Gemma 3 goes multimodal ! The 4, 12, and 27 billion parameter models can process both images and text , while the 1B variant is text only .
Gemma 3 comes with day zero support in transformers . All you need to do is install transformers from the stable release of Gemma 3.
Gemma 3 ships with day zero support in mlx-vlm , an open source library for running vision language models on Apple Silicon devices, including Macs and iPhones
The input context window length has been increased from Gemma 2’s 8k to 32k for the 1B variants, and 128k for all others. As is the case with other VLMs (vision-language models), Gemma 3 generates text in response to the user inputs, which may consist of text and, optionally, images. Example uses include question answering, analyzing image content, summarizing documents, etc.
While these are multimodal models, one can use it as a text only model (as an LLM) without loading the vision encoder in memory. We will talk about this in more detail later in the inference section.
In this section, we will cover the technical details that lead to these enhancements. It is interesting to start with the knowledge of Gemma 2 and explore what was necessary to make these models even better. This exercise will help you think like the Gemma team and appreciate the details!
Welcome Gemma 3: Google's all new multimodal, multilingual, long context open LLM. Covered by Hugging Face on March 12, 2025.