Gemma 3
Recommend & Upvote | 5 I am a co-founder
Gemma 3 is Google's latest multimodal AI model, capable of handling text, images, and video. Available in sizes ranging from 1B to 27B parameters, it supports a 128K context window and over 140 languages. The suite also features ShieldGemma 2, an enhanced safety module.
Discover Gemma 3, Google's newest family of models designed for creating multimodal AI applications! This version marks a significant upgrade from earlier Gemma models, introducing video understanding and a vastly expanded context window.
Key Features:
🖼️ Multimodal Capabilities: Works seamlessly with text, images, and short videos.
🧠 Flexible Sizes: Choose from 1B, 4B, 12B, and 27B parameter versions.
↔️ 128K Context Window: A substantial improvement, enabling the processing of much larger amounts of information.
🌍 Multilingual Support: Out-of-the-box compatibility with over 35 languages, pretrained on more than 140.🛠️ Integration: Works with Hugging Face Transformers, Ollama, JAX, Keras, PyTorch, Unsloth, vLLM, and Gemma.cpp.
🛡️ Safety Features: Includes ShieldGemma 2, a dedicated 4B model for image safety classification.
⚡ Optimized Performance: Tailored for NVIDIA GPUs, Google Cloud TPUs, and AMD GPUs.
Gemma 3 is a testament to the rapid evolution of multimodal AI technology.
Let’s dive in and explore its potential in Google AI Studio!
Free options
Tags : Open Source , Artificial Intelligence , Development
Log in or create a new account to comment