Gemma 2B

Gemma 2B: Google's lightweight, open-weight language model (2 billion parameters) delivers best-in-class performance for efficient on-device and cloud deployment.

Gemma 2B is Google's highly efficient, open-weight language model, built directly from the research and technology behind the Gemini models. With 2 billion parameters, this decoder-only LLM is engineered for maximum performance at a minimal size, making it ideal for resource-constrained environments like laptops and edge devices. It supports an 8192-token context length and excels across core text-generation tasks: question answering, summarization, and reasoning. The model’s small footprint and open licensing democratize access to state-of-the-art AI, enabling rapid fine-tuning and flexible deployment across various hardware platforms.

https://huggingface.co/google/gemma-2b