Google’s new open AI tools for everyday tech

Google just dropped a set of free AI models called Gemma 4, built from the same tech behind its premium Gemini 3 systems. Instead of keeping them locked up, the company made these models public under a flexible license that lets anyone tweak and use them however they want. That’s a big deal because most cutting-edge AI tools stay behind closed doors, making this a rare chance for developers to play with high-end tech without paying fees. The lineup includes four versions that fit different needs. The smallest ones, with 2 billion and 4 billion settings, work on phones and basic computers. The bigger ones need more power: a 26-billion version and a 31-billion version that can handle complex tasks. Google says these models punch above their weight, ranking high on text tests even when compared to much larger systems. They also handle video, images, and speech, which means they could power apps like instant translators or smart assistants.One cool feature is offline coding. You can write or debug programs without an internet connection, which is useful for places with weak signals or strict data rules. The models also speak over 140 languages, making them handy for global projects. Since Google released them under an open license, anyone can adjust the code, run it on their own servers, or even sell their own versions. That’s a shift from past releases, which came with tighter restrictions. Not everyone will benefit, though. Running the bigger models still needs serious hardware, so regular users might find them too slow. And while the open license is great for tech experts, it could lead to misuse if people modify the models without proper checks. Still, this move could push more innovation, since developers now have a powerful toolkit to experiment with.