Google unveils Gemma, bringing AI power at the fingertips of developers

Google has joined the AI race with Gemeni, and they plan to release Gemini Business and Gemini Enterprise soon.

However, the company has released an open model called Gemma, that everyone should be able to use.

Google Gemma lets developers make their AI-powered apps

According to the latest Google blog post, Gemma is a lightweight open model that uses the same technology as its bigger brother Gemini.

This allows Gemma to achieve the best performance compared to other open models. This model will closely follow Google’s AI principles and the company is also releasing a new Responsible Generative AI Toolkit that will help developers to focus on building safe and responsible AI applications.

Regarding compatibility, you can use this Gemma with various frameworks including Keras 3.0, PyTorch, JAX, and Hugging Face Transformers.

As for supported hardware, you should be able to run this model on just about anything ranging from a PC and mobile to IoT and cloud. As for hardware, Google has partnered with NVIDIA for better optimization on NVIDIA GPUs.

Lastly, the Gemma is optimized for Google Cloud and thanks to Vertex AI and an MLOps toolset it allows seamless deployment.

You can start using Gemma today thanks to free access in Kaggle, and first-time Google Cloud users will get $300 in credits. Researchers can apply for Google Cloud credits and get up to $500,000 in credits for projects.

It’s also worth mentioning that Google has implemented Gemini in the Chrome search bar, and if you haven’t tried this, this might be the right time to do so.