Google Gemini 3.1 Pro First Impressions: The ‘Deep Think Mini’ With Adjustable Thinking On Demand

Over the past three months, Google’s Gemini 3 Pro has held its ground as one of the most capable edge models available. But in the fast-moving world of AI, three months is a lifetime – and competitors are not standing still.

Earlier today, Google released Gemini 3.1 Proan update that brings a key innovation to the company’s workhorse power model: three levels of adjustable thinking that effectively make it a lightweight version of Google’s specialized Deep Think reasoning system.

The release marks the first time Google has released a "point one" update to the Gemini model, signaling a shift in the company’s release strategy from periodic full release releases to more frequent incremental upgrades. More importantly for enterprise AI teams evaluating their set of models, 3.1 Pro’s new three-tier thinking system—low, medium, and high—gives developers and IT leaders a single model that can dynamically scale their reasoning efforts, from quick answers for routine queries to multi-minute deep thinking sessions on complex problems.

The model is now distributed in preview on the Gemini API via Google AI StudioGemini CLI, the Google Antigravity agent development platform, Vertex AI, Gemini Enterprise, Android Studio, the Gemini user app, and NotebookLM.

The “Deep Think Mini” Effect: Adjustable Reasoning on Demand

The most significant feature in Gemini 3.1 Pro isn’t a single comparison number—it’s the introduction of a three-tier thinking level system that gives users precise control over how much computational effort the model invests in each response.

Gemini 3 Pro offers only two thinking modes: low and high. The new 3.1 Pro adds a mid setting (similar to the previous high) and, critically, revises what "high" means. When set to high, the 3.1 Pro behaves like a "a mini version of Gemini Deep Think" — the company’s specialized reasoning model that was just updated last week.

The implications for enterprise deployment can be significant. Instead of routing requests to different specialized models based on task complexity—a common but labor-intensive model—organizations can now use a single model endpoint and adjust the depth of reasoning based on the task at hand. Routine document summarization can be run low-thinking with fast response times, while complex analytical tasks can be elevated to high-level thinking for Deep Think-caliber reasoning.

Benchmark performance: More than double the reasoning over the 3 Pro

Google’s published benchmarks tell a story of dramatic improvement, particularly in areas related to reasoning and agentic capabilities.

included ARC-AGI-2a benchmark that evaluates the model’s ability to solve new abstract reasoning patterns, score 3.1 Pro 77.1% — more than double the 31.1% achieved by Gemini 3 Pro and well ahead of Anthropic’s Sonnet 4.6 (58.3%) and Opus 4.6 (68.8%). This score also eclipses OpenAI’s GPT-5.2 (52.9%).

Profits spread everywhere. included Humanity’s final testrigorous benchmark for academic reasoning, the 3.1 Pro scored 44.4% without tools, compared to 37.5% for the 3 Pro and ahead of the Claude Sonnet 4.6 (33.2%) and Opus 4.6 (40.0%). included GPQA Diamondscientific knowledge assessment, the 3.1 Pro reached 94.3%, surpassing all listed competitors.

Where the results become particularly relevant for enterprise AI teams is in agency benchmarks—assessments that measure how well models perform when given tools and multi-step tasks, the kind of work that increasingly defines production AI deployments.

included Terminal Bench 2.0which evaluates agent terminal coding, the 3.1 Pro scored 68.5% compared to 56.9% for its predecessor. included MCP atlasbenchmark measuring multi-step workflows using the Model Context Protocol, the 3.1 Pro hit 69.2%—a 15-point improvement over the 3 Pro’s 54.1% and nearly 10 points ahead of Claude and GPT-5.2. And on BrowseCompwhich tested the agent’s ability to search the web, the 3.1 Pro achieved 85.9%, beating the 3 Pro’s 59.2%.

Why Google chose version « 0.1 » — and what it means

The release decision itself is remarkable. Previous releases of Gemini followed a pattern of dated previews – multiple 2.5 previews, for example, before reaching general availability. The choice to label this update as 3.1 rather than another 3 Pro preview suggests that Google sees the improvements as substantial enough to warrant a version bump, while "point one" the frame creates expectations that this is an evolution, not a revolution.

Google’s blog post states that 3.1 Pro builds directly on lessons from the Gemini Deep Think series, incorporating techniques from earlier and newer versions. The benchmarks strongly suggest that reinforcement learning played a central role in the gains, especially in tasks such as the ARC-AGI-2, coding benchmarks, and agent evaluations—precisely the domains where RL-based learning environments can provide clear reward signals.

The model is being released in preview rather than a general launch, with Google saying it will continue to advance areas like agent workflows before moving to full GA.

Competitive implications for your enterprise AI stack

For IT decision makers evaluating edge model vendors, the release of Gemini 3.1 Pro should not only make them rethink which models to choose, but also how to adapt to such a rapid pace of change for their own products and services.

The question now is whether this release provokes a response from competitors. The initial release of the Gemini 3 Pro last November set off a wave of model releases in both proprietary and open ecosystems.

With 3.1 Pro regaining the benchmark lead in several critical categories, the pressure is on Anthropic, OpenAI, and the open-weight community to respond—and in the current AI landscape, that response is likely to be measured in weeks, not months.

Availability

Gemini 3.1 Pro is now available in preview through the Gemini API in Google AI Studio, Gemini CLI, Google Antigravity, and Android Studio for developers. Enterprise customers can access it through Vertex AI and Gemini Enterprise. Users of Google AI Pro and Ultra plans can access it through the Gemini app and NotebookLM.

Technology

#Google #Gemini #Pro #Impressions #Deep #Mini #Adjustable #Thinking #Demand