Gemini 3.1 Pro: A smarter model for your most complex tasks
Google has just announced the release of Gemini 3.1 Pro, a major upgrade to its core AI intelligence designed specifically to tackle complex challenges across science, research, and engineering. Built on the foundation of the Gemini 3 series, this new model is tailored for tasks that require advanced reasoning and where simple answers simply aren’t enough.
A standout feature of Gemini 3.1 Pro is its massive leap in logic capabilities. On the rigorous ARC-AGI-2 benchmark—which evaluates a model's ability to solve entirely new logic patterns—Gemini 3.1 Pro achieved a verified score of 77.1%. This marks more than double the reasoning performance of its predecessor, Gemini 3 Pro.
For university students, researchers, and developers, this enhanced intelligence opens up exciting practical applications. The model excels at synthesizing data into single views, building interactive designs (such as coding immersive 3D experiences with hand-tracking), and generating website-ready animated SVGs directly from text prompts. It can even perform "creative coding," such as mode of an image or the creative tone of a document to design a sleek, functional website interface reflecting its themes. Find out more in Google’s blog post.
Lyria 3: Next level music generation
Google also just released Lyria 3, its latest music general model. It has the capability to create music in many different genres with vocals in many different languages. Music that is produced has an imperceptible "watermark" that identifies it as AI-generated. The service is respectful of copyright and the artistic expression of artists. When a prompt names a particular artist, Gemini will create a track that shares a similar style or mood, but not directly imitate the artist. You can even upload an image and have Gemini create music to match the image's mood or context. Find out more in Google's blog post.
How to access these models
Both are available now to the campus community at https://gemini.google.com. Gemini 3.1 Pro is available in the model dropdown on the left side of the chatbox with the options of Fast, Thinking, and Pro. To use Lyria 3, select "Create music" in the "Tools" dropdown menu. Gemmini 3.1 Pro is also available to developers via the Gemini API in Google AI Studio, Gemini CLI, Vertex AI, and Antigravity.