Share your Gemma 4 builds or the model variants you’re training in the replies below!

Google released Gemma 4 in April 2025 as its most capable open-weight model family to date, built on the same research as Gemini 3 and licensed under Apache 2.0 for unrestricted commercial use, fine-tuning, and redistribution. The family spans multiple sizes (E2B, E4B, 26B MoE, and 31B dense) and introduces native multimodal support — including vision, audio, and text — alongside chain-of-thought reasoning, function calling, and context windows up to 256K tokens across 140+ languages. Google's official launch post on X invited the developer community to share their Gemma 4 builds and fine-tuned model variants, reflecting the project's emphasis on community-driven customization and the broader "Gemmaverse" ecosystem.

We love seeing what you’ve built with Gemma 4, the open model family that we released last week. Here are a few fun examples, described by t…
Google just casually disrupted the open-source AI narrative…
Gemma 4 on AI Gateway
Lots of love for Gemma 4! Team just told me it’s already had 10M+ downloads since last week’s launch. Gemma models have now been downloaded …
Gemma 4 31B brings dense multimodal reasoning to Together AI. Try Now: http://www.together.ai/models/gemma-4-31b
Having a model like Gemma 4, which is perfectly adequate for everyday use in many cases, runs locally, is free, and secure, still feels unre…

Source: model-releases

Related