Google has commenced the rollout of Gemini 3.1 Pro, an enhanced version of its flagship AI model designed for intricate reasoning tasks, across its consumer, developer, and enterprise products. This update introduces capabilities aimed at scenarios requiring multi-step reasoning and synthesis, expanding the model’s application in various sectors.
The model is currently available in preview for developers via the Gemini API in Google AI Studio. It can also be accessed through Gemini CLI, Google Antigravity, and Android Studio. Enterprise clients can utilize it through Vertex AI and Gemini Enterprise, while consumers can engage with it via the Gemini app and NotebookLM.
Google has characterized Gemini 3.1 Pro as a significant advancement in what it terms core reasoning for the Gemini 3 series, specifically tailored for situations where direct responses fall short. The company emphasized that this model is designed to tackle more complex and nuanced problems.
In terms of performance, Google cited a verified score of 77.1% on the ARC-AGI-2 benchmark, which assesses a model’s ability to solve unfamiliar logic patterns. This achievement is reported to be more than double the reasoning performance of the previous Gemini 3 Pro model.
The launch follows a recent upgrade to Gemini 3 Deep Think, aimed at enhancing applications in science, research, and engineering. Google referred to Gemini 3.1 Pro as the “upgraded core intelligence” that underpins these advancements, indicating a commitment to elevating AI capabilities across various domains.
For software developers, the early access to Gemini 3.1 Pro integrates the model into multiple components of Google’s development stack. The Gemini API in AI Studio provides a standardized approach for application integration, while Gemini CLI offers a command-line interface. The model’s presence in Android Studio allows for easier incorporation into mobile development workflows, and Google Antigravity serves as a platform for testing agent-focused applications.
Google’s rollout of Gemini 3.1 Pro across its consumer and developer products is framed as an effort to integrate the latest model improvements into everyday software usage. The company has shared various examples highlighting the model’s capabilities in design, coding, and data visualization, showcasing its practical applications.
One instance described the generation of animated SVG graphics from a text prompt, resulting in “website-ready, animated SVGs” that emphasize code-based outputs rather than pixel-based images. Another example illustrated the configuration of a public telemetry stream used to create “a live aerospace dashboard” visualizing the orbit of the International Space Station.
In a creative design demonstration, Gemini 3.1 Pro produced code for a 3D starling murmuration, incorporating features such as hand-tracking and audio that adjusted according to the simulated birds’ movement. Furthermore, the model created a modern personal portfolio inspired by Emily Brontë’s “Wuthering Heights,” aligning the interface with the novel’s tone.
For consumers, the rollout of Gemini 3.1 Pro in the Gemini app includes higher usage limits for users on the Google AI Pro and Ultra plans. Access to NotebookLM is similarly restricted to Pro and Ultra subscribers, reflecting a trend in the AI market where advanced model upgrades primarily benefit paying customers and developers initially.
Google has launched Gemini 3.1 Pro in preview as it continues validation and development, aiming for “ambitious agentic workflows” as a focal point. The company anticipates general availability of the model in the near future, citing user feedback and the rapid progress since the launch of Gemini 3 Pro in November as key motivators for this update.
As the AI landscape evolves, Google’s introduction of Gemini 3.1 Pro underscores its ongoing commitment to enhancing model capabilities, aiming to address increasingly complex user needs while ensuring that these advanced technologies remain accessible across its diverse product offerings.
See also
AI Models Face Off: DeepSeek-R1 Achieves 98% Excellence in Ovarian Cancer Care Evaluation
Germany”s National Team Prepares for World Cup Qualifiers with Disco Atmosphere
95% of AI Projects Fail in Companies According to MIT
AI in Food & Beverages Market to Surge from $11.08B to $263.80B by 2032
Satya Nadella Supports OpenAI’s $100B Revenue Goal, Highlights AI Funding Needs
