Google publishes 1,134 models on its google organization on Hugging Face<\/a>. For local LLM users, the center of it is Gemma<\/strong>, the open model family built by Google DeepMind. The organization card describes Gemma as “a family of lightweight, state-of-the-art open models from Google, built from the same research and technology used to create the Gemini models”.<\/p>\n Besides Gemma, the organization hosts many models for other purposes, such as MedGemma for medicine, TranslateGemma for translation, Magenta RealTime for music and TimesFM for time-series forecasting. The table below lists only the generative models this site covers.<\/p>\n The Gemma 4 and DiffusionGemma model cards give the license as Apache-2.0<\/strong> (linking to the Gemma 4 license page). TranslateGemma from January 2026 uses the earlier Gemma license<\/strong> ( Sources: the google organization card<\/a>; model cards of google\/gemma-4-31B<\/a>, google\/gemma-4-12B-it-qat-q4_0-gguf<\/a> and google\/diffusiongemma-26B-A4B-it<\/a> (all as of 2026-09-26).<\/em><\/p>\n Local Model Watch has not published an article on Google’s models yet. The model list below is collected by code from Hugging Face every day. Part of our publisher index<\/a>; model families link to their family pages<\/a>.<\/p>\nWhat It Releases<\/h2>\n
\n
Licenses<\/h2>\n
gemma<\/code>), and models such as TimesFM use non-commercial licenses. Terms differ from model to model within the same organization, so check the license column in the table and each model’s license text. This summary is not legal advice.<\/p>\nRunning Them Locally<\/h2>\n
\n
Our Coverage and Data<\/h2>\n
At a Glance<\/h2>\n