✨ Introducing Gemma 3n, available in early preview today. The model uses a cutting-edge architecture optimized for mobile on-device usage. It brings multimodality, super fast inference, and more.

Key features include: -Expanded multimodal understanding with video and audio input, alongside text and images -Developer-friendly sizes: 4B and 2B (and many in between!) -Optimized on-device efficiency for 1.5x faster response on mobile compared to Gemma 3 4B

Build live, interactive apps and sophisticated audio-centric experiences, including real-time speech transcription, translation, and rich voice-driven interactions ↓

We’re working to roll out full multimodal features, and are collaborating with partners to bring Gemma 3n to the open-source community in the coming weeks. Explore the model’s text and image capabilities in AI Edge, and text capabilities in Google AI Studio:

Our speech-to-text models are the most accurate on the market with top rankings across industry benchmarks. - The highest accuracy rates—up to 95% - Up to 30% fewer hallucinations than other leaders - Low latency—63 minutes converts in 35 seconds Try via API for free today 👇

google shipping mobile first functionality? W

Google won! They are changing the game

How can we use this? And is this the only feature from the new ones which is available right now?

it refuses image and audio inputs

N

Cutting edge is the most horrible two words of the AI Gen age

Gemma 3n’s focus on mobile and fast inference is a smart move. Running multimodal models directly on devices opens up a lot of possibilities.