Gemini 3.1 Pro Brings "Double the Reasoning Performance" to Complex Tasks

Google is pushing deeper into advanced reasoning. The company released Gemini 3.1 Pro, an upgraded model designed for tasks where "a simple answer isn't enough," according to the announcement.

The model achieves a verified score of 77.1% on ARC-AGI-2, a benchmark that tests a model's ability to solve entirely new logic patterns. That's more than double the reasoning performance of Gemini 3 Pro, the company said.

"3.1 Pro is a smarter, more capable baseline for complex problem-solving," Google wrote in announcing the release on February 19, 2026.

The model is rolling out across Google's ecosystem: developers can access it via the Gemini API in Google AI Studio, Gemini CLI, Antigravity, and Android Studio. Enterprises get it through Vertex AI and Gemini Enterprise. Consumers can find it in the Gemini app and NotebookLM (Pro and Ultra plans only for the latter).

The focus is on practical complexity. Google showcased examples: generating animated SVGs from text prompts, building live aerospace dashboards that visualize the International Space Station's orbit by connecting to public telemetry streams, and creating 3D starling murmurations with hand-tracking and generative audio.

"This improved intelligence can help in practical applications—whether you're looking for a clear, visual explanation of a complex topic, a way to synthesize data into a single view, or bringing a creative project to life," the company noted.

Google is releasing 3.1 Pro as a preview—"to validate these updates and continue to make further advancements in areas such as ambitious agentic workflows before we make it generally available soon."