Google Launched Gemma - A Next Generation Open Model for Gemini

Gemma is a family of lightweight, state-of-the-art open models built from the same research and technology used to create the Gemini models. It is developed by Google DeepMind team. Gemma is inspired by Gemini, and the name reflects the Latin gemma, meaning precious stone.

Google released model weights in two sizes: Gemma 2B and Gemma 7B. Each size is released with pre-trained and instruction-tuned variants. It is responsible by design, it pre-trains curated data and make it safe for AI development. Its comprehensive evaluations and transparent reporting unveil model limitations to adopt a responsible approach for each use case.

A new Responsible Generative AI Toolkit provides guidance and essential tools for creating safer AI applications with Gemma. This toolkit will investigate Gemma's behavior to cover risk and mitigation techniques to address safety, privacy, fairness, and accountability.

It also provides toolchains for inference and supervised fine-tuning (SFT) across all major frameworks: JAX, PyTorch, and TensorFlow through native Keras 3.0. It is optimized across multiple AI hardware platforms ensures industry-leading performance, including NVIDIA GPUs and Google Cloud TPUs.

Gemma models can run on laptop, workstation, Google Cloud, Google Kubernetes Engine and Vertex AI.