Google has launched Gemini 3.1 Pro, an advanced AI model that builds on the Gemini 3 series. This new version targets users in science, research, engineering, creative technology, and enterprise development who require high-level reasoning and complex problem-solving capabilities. The update is available immediately as a preview for Google AI Pro and Ultra plan users through the Gemini app and NotebookLM, and also accessible to developers and businesses via platforms like AI Studio, Vertex AI, Gemini Enterprise, Gemini CLI, and Android Studio.
Gemini 3.1 Pro is here.
— Google DeepMind (@GoogleDeepMind) February 19, 2026
We’ve significantly improved the model’s overall intelligence so it can solve tougher problems. 🧵 pic.twitter.com/upyJv1p7MA
Gemini 3.1 Pro introduces significant improvements in logical reasoning, achieving a 77.1% verified score on the ARC-AGI-2 benchmark, which is over twice the result of the previous 3 Pro model. The model can generate scalable, code-based SVG animations, synthesize complex systems into intuitive dashboards, and create interactive, sensory-rich experiences, such as live 3D simulations with real-time hand tracking and generative audio. It also demonstrates the ability to translate abstract literary themes into functional, visually compelling interfaces, setting it apart from earlier iterations and competitors.
The rollout of Gemini 3.1 Pro reflects Google’s ongoing efforts to deliver more powerful AI tools to its user base. The company has positioned this release as a step toward supporting more ambitious, agentic workflows for professionals and developers, emphasizing continuous feedback and rapid iteration since the initial Gemini 3 Pro launch.
