Google Rolls Out Gemini 3.1 Pro
Google is rolling out Gemini 3.1 Pro, an updated AI model with improved complex reasoning and reliability for enterprise data teams. The AI assistant is now built directly into the Chrome browser, offering access to AI-powered SQL generation and data exploration. Google has also introduced tiered pricing plans targeting professional and enterprise analytics use cases.
- Gemini 3.1 Pro demonstrates significantly improved reasoning, scoring 77.1% on the ARC-AGI-2 benchmark, which evaluates a model's ability to solve new logic patterns—more than double the performance of Gemini 3 Pro. It also shows enhanced capabilities in software engineering, achieving an 80.6% score on the SWE-Bench benchmark for resolving real-world GitHub issues. - For developers, Gemini 3.1 Pro is accessible in preview via the Gemini API in Google AI Studio, Vertex AI, and Gemini Enterprise. It includes a 1 million token context window and a specific endpoint, `gemini-3.1-pro-preview-customtools`, optimized for agentic workflows that use custom tools alongside bash. - The model introduces a new "medium" thinking level, designed to offer a better balance between cost, performance, and speed for complex tasks. This is part of a broader effort to improve token efficiency and reduce latency in agentic workflows. - Pricing for the preview of Gemini 3.1 Pro is set at $2.00 per 1 million input tokens and $12.00 per 1 million output tokens for prompts up to 200k tokens. This positions it competitively against other frontier models like Anthropic's Claude Opus 4.6 and OpenAI's GPT-5.2. - Within Google's data analytics ecosystem, Gemini is being integrated into BigQuery and Looker to assist with the entire data lifecycle. This includes natural language-based query generation, creation of visualizations, and AI-driven data summarization in Looker Studio. - The integration with BigQuery enables natural language to generate and explain SQL and Python code, recommend data partitioning for query optimization, and support low-code visual data pipeline development. This aims to lower the technical barrier for data analysis and increase productivity for data professionals. - Gemini 3.1 Pro's advanced multimodal capabilities allow it to process and reason across text, images, audio, video, and code from a single prompt. This enables complex use cases such as analyzing hour-long videos or entire code repositories. - For enterprise use, the model is optimized for reliability and factual consistency, with features designed for structured domains like finance and spreadsheet automation. Its integration with Google Workspace allows for high-context task completion using @ tagging to ground requests in user data from Docs and Gmail.
