On June 28th, Google announced the release of Gemma 2 large language model for global researchers and developers. It is stated that Gemma 2 has two parameter sizes of 9 billion (9B) and 27 billion (27B), with higher performance and inference efficiency than the first generation, and significant security improvements built in. Google claims that the performance of Gemma 2 27B is more competitive than products with twice its size, and the performance of 9B is also in the leading position among similar products, superior to Llama 3 8B and other open models.
谷歌Gemma 2发布，与第一代相比性能更高
Google has released Gemma 2, which has higher performance than the first generation.
The translation is provided by third-party software.
The above content is for informational or educational purposes only and does not constitute any investment advice related to Futu. Although we strive to ensure the truthfulness, accuracy, and originality of all such content, we cannot guarantee it.