According to Odaily, Google has announced the global release of its Gemma 2 language model for researchers and developers. The Gemma 2 model comes in two parameter sizes, 9 billion (9B) and 27 billion (27B). Compared to its predecessor, Gemma 2 offers improved performance, higher inference efficiency, and significant security enhancements.
Google claims that the performance of the 27B Gemma 2 model is more competitive than similar products that are twice its size. The 9B model also leads in its category, outperforming the Llama 3 8B and other open models. The release of Gemma 2 signifies Google's commitment to providing advanced tools for language research and development, while also prioritizing security and efficiency.