Gemini 3.1 Pro Preview

• 🧠 Google's frontier reasoning model, built on Gemini 3 Pro.
• 📏 1M-token context window spanning multimodal inputs.
• 👁️ Handles text, audio, images, video, PDFs, and code.
• 🔧 Function calling, web search, and agentic tool orchestration.
• 🎯 Tuned for software engineering and agentic reliability.
• 🆕 Preview released February 2026 to validate updates.
• 🏢 Available via Gemini API, AI Studio, and Google Cloud.

Google

Google is an American multinational technology corporation and one of the world's most valuable brands. A subsidiary of parent company Alphabet Inc., Google operates across search, cloud computing, consumer electronics, and artificial intelligence. Its…

Gemini 3.1 Pro Preview is Google's latest flagship frontier model, positioned as the company's most advanced reasoning system for complex, multi-step problems. According to its DeepMind model card, it is based directly on Gemini 3 Pro and comprehends massively multimodal inputs — text, audio, images, video, PDFs, and entire code repositories — within a 1M-token context window. It supports reasoning, vision, function calling, web search, and audio input.

Compared with its same-family predecessor Gemini 3 Pro (whose preview endpoint Google discontinued in March 2026), the DeepMind model card describes 3.1 Pro as based directly on Gemini 3 Pro, refining the series for stronger reasoning and more reliable agentic workflows. Google's documentation positions it for software engineering behavior, usability, and tool-driven, multi-step tasks.

Within Google's broader 2026 lineup, the Pro tier sits above lighter siblings such as Gemini 3 Flash Preview and the later Gemini 3.5 Flash, which prioritize speed and lower cost. Google released 3.1 Pro in preview to validate updates and advance ambitious agentic workflows before general availability, shipping it across developer tools including the Gemini API, AI Studio, and Google Cloud.