Google’s Gemini 3.1 Pro Doubles Reasoning Performance, Now Available

Google has released Gemini 3.1 Pro, the latest iteration in its Gemini series of AI models, boasting significant improvements in reasoning capabilities and performance across various benchmarks. Announced on February 19, 2026, Gemini 3.1 Pro is now available in preview for developers and businesses through Gemini Enterprise and Vertex AI.

What is Gemini 3.1 Pro?

Gemini 3.1 Pro is Google’s most advanced model for complex tasks, designed to comprehend vast datasets and challenging problems from multimodal information sources, including text [Google DeepMind]. It represents a “noticeably smarter, more capable baseline for complex problem-solving” according to Google [Google Cloud Blog].

Key Performance Improvements

Testing reveals Gemini 3.1 Pro achieves more than double the reasoning performance of the previous Gemini 3 Pro model on the ARC-AGI-2 benchmark, scoring 77.1% [ZDNet]. This benchmark assesses a model’s ability to solve logic puzzles with entirely new patterns.

Compared to other leading models, Gemini 3.1 Pro demonstrates superior performance:

• Gemini 3.1 Pro: 77.1%
• GPT-5.2: 52.9%
• Claude Opus 4.6: 68.8%

Google outperformed competitors across 12 of 19 benchmarks [ZDNet]. In the GPQA Diamond test, which evaluates scientific knowledge, Gemini 3.1 Pro scored 94.3%, surpassing GPT-5.2 (92.4%) and Claude Opus 4.6 (91.3%) [Mashable].

Accessing Gemini 3.1 Pro

Gemini 3.1 Pro is currently available in preview through several platforms:

• Vertex AI
• Gemini Enterprise
• Gemini CLI
• Android Studio
• Google AI Studio
• Notebook LM (requires Google AI Pro or AI Ultra subscription)
• Gemini App (free users have limited access, paid users have higher usage limits)

Users can switch to Gemini 3.1 Pro within the Gemini app by selecting the model name in the prompt box, which will display a description: “Advanced math and code with 3.1 Pro” [Mashable].

Real-World Applications

Gemini 3.1 Pro is designed to tackle complex challenges, enabling users to:

• Visualize complex topics
• Synthesize data into a single view
• Solve problems requiring deep context and planning

Early adopters, such as JetBrains, have reported up to a 15% improvement in quality and faster processing speeds with Gemini 3.1 Pro [Google Cloud Blog]. Databricks is also integrating Gemini 3.1 Pro to empower teams building and governing GenAI applications [Google Cloud Blog].