Google Unleashes Gemma 4: A Multimodal AI Powerhouse Redefining Language and Vision Tasks
Google has launched Gemma 4, a multimodal AI model handling text, images, and audio. It features a 256K token context, supports 140+ languages, and uses hybrid dense/MoE architectures. Available in four sizes from E2B to 31B, it's deployable from mobile to server.
Read More