Models·1 min read·Google AI Blog

Google Gemini 2.0: The Model That Unifies Everything

Google unveils Gemini 2.0 with advanced multimodal capabilities and deep integration across all Google services.

Google Gemini 2.0: The Model That Unifies Everything
Share:

Gemini 2.0

Google presented Gemini 2.0, a new generation of AI models that seamlessly combines text, images, video, and audio. The new model integrates directly into Gmail, Docs, Search, and Android.

Project Astra, a Gemini 2.0-powered AI assistant, offers a real-time multimodal conversation experience.

Want AI news before everyone else?

The morning's most important AI stories, straight to your inbox. No fluff.

← Back to News

Related Articles

Shieldstral Mistral's 3B open-weights safety guard — you write the policy at runtime policy.txt "Flag content that promotes violence or self-harm." plain language, at inference 3B Apache 2.0 open weights calibrated score 0.97 BLOCK one forward pass, yes/no logits 84.9 F1 text · 83.8 F1 multimodal · 12 languages · runs on one 16 GB GPU BITSMINDS.COM
Models

Mistral's Shieldstral: A 3B Guard That Reads Your Policy

MAI PLAYGROUND · HIDDEN EARLY ACCESS MAI-Realtime does not take turns. It listens and speaks together. OUT · SPEAKS IN · LISTENS hears you barge in never waits for a pause SAME INSTANT VOICES Victoria · Grant LANGUAGES 17, switched mid-sentence STATUS unannounced, no pricing BITSMINDS.COM
Models

Microsoft's MAI-Realtime Aims at OpenAI's Voice Slot

QWEN3.8-MAX Weights next week.
Models

Alibaba Opens Qwen3.8-Max to the World — and Says the Weights Come Next Week