Google Releases Gemini 3.1 Pro With Advanced Reasoning Capabilities; Pomelli Adds AI Photoshoot Feature

Google has introduced Gemini 3.1 Pro, the latest AI model in its Gemini 3 AI series. Google says this version is designed to handle more advanced reasoning tasks and complex problem solving compared to earlier releases. Alongside this update, Google Labs has added a new Photoshoot feature to Pomelli, its experimental AI marketing assistant, allowing users to generate campaign ready images from a single product photo.

Gemini 3.1 Pro Availability

Gemini 3.1 Pro is rolling out across Google’s developer, enterprise, and consumer platforms. Developers can access it through the Gemini API in Google AI Studio, Gemini CLI, Android Studio, and Google Antigravity, which is Google’s agent development platform.

Enterprise users can deploy the model through Vertex AI and Gemini Enterprise products. Individual users can use Gemini 3.1 Pro in the Gemini app and NotebookLM, with higher usage limits available for Google AI Pro and Ultra subscribers.

This broad rollout reflects Google’s strategy of integrating advanced AI into productivity tools, enterprise systems, and development environments.

Benchmark Performance and Reasoning Improvements

Google reports that Gemini 3.1 Pro delivers stronger reasoning performance than its predecessor. On the ARC AGI 2 benchmark, which measures an AI model’s ability to solve new and unfamiliar logic problems, Gemini 3.1 Pro achieved a verified score of 77.1 percent.

For comparison, earlier Gemini versions and several competing large language models have reported lower scores on similar evaluations. In another academic reasoning evaluation known as Humanity’s Last Exam, Gemini 3.1 Pro also showed improved performance over previous Gemini models.

These benchmark results indicate better generalization ability, meaning the model can interpret and solve new types of problems more effectively.

Improvements in Creative Output

Google has also demonstrated improvements in creative generation tasks. When given identical prompts, Gemini 3.1 Pro generated more detailed and structured Scalable Vector Graphics compared to Gemini 3 Pro.

This enhancement is particularly useful for developers and designers who rely on precise visual outputs. The model continues to support multimodal processing, allowing it to work with both text and visual inputs.

Pomelli’s New AI Photoshoot Feature

In a separate update, Google Labs introduced a Photoshoot feature within Pomelli, its AI powered marketing assistant.

The feature allows users to upload a single product image and generate multiple promotional visuals suitable for advertisements, e commerce listings, and social media campaigns. The tool can create different backgrounds, angles, lighting effects, and promotional layouts based on the original reference image.

For example, a single smartphone image can be transformed into multiple marketing banners with varied styles and text overlays. This reduces the need for traditional studio photography and can help businesses create marketing assets more efficiently.

Availability of Photoshoot

The Photoshoot feature is currently rolling out to users in the United States, Canada, Australia, and New Zealand. Google has positioned it as a free experimental feature during its early phase.

Gemini 3.1 Pro represents a measurable improvement in reasoning performance within Google’s AI ecosystem, supported by stronger benchmark results and wider platform integration. At the same time, Pomelli’s new Photoshoot feature demonstrates Google’s focus on practical AI tools for businesses and marketers. Together, these updates show how Google is expanding both advanced AI research capabilities and real world applications.