google/gemma-3-27b-it

Gemma 3 is a family of lightweight, state-of-the-art open models from Google DeepMind, built from the same research and technology used to create the Gemini models. The 27 billion parameter instruction-tuned variant (gemma-3-27b-it) is multimodal, handling text and image input with a large 128K context window, and generates text output. It offers multilingual support in over 140 languages and is well-suited for a variety of text generation and image understanding tasks, including question answering, summarization, and reasoning.