Google Gemini 3.1 Pro doubles reasoning score, beats rivals in key AI benchmarks

Google announces Gemini 3.1 Pro, says it's better at complex problem-solving

Another day, another Google AI model. Google has really been pumping out new AI tools lately, having just released Gemini 3 in November. Today, it's bumping the flagship model to version 3.1. The new Gemini 3.1 Pro is rolling out (in preview) for developers and consumers today with the promise of better problem-solving and reasoning capabilities. Google announced improvements to its Deep Think tool last week, and apparently, the "core intelligence" behind that update was Gemini 3.1 Pro. As usual, Google's latest model announcement comes with a plethora of benchmarks that show mostly modest improvements. In the popular Humanity's Last Exam, which tests advanced domain-specific knowledge, Gemini 3.1 Pro scored a record 44.4 percent. Gemini 3 Pro managed 37.5 percent, while OpenAI's GPT 5.2 got 34.5 percent. Google also calls out the model's improvement in ARC-AGI-2, which features novel logic problems that can't be directly trained into an AI. Gemini 3 was a bit behind on this evaluation, reaching a mere 31.1 percent versus scores in the 50s and 60s for competing models. Gemini 3.1 Pro more than doubles Google's score, reaching a lofty 77.1 percent. Google has often gloated when it releases new models that they've already hit the top of the Arena leaderboard (formerly LM Arena), but that's not the case this time. For text, Claude Opus 4.6 edges out the new Gemini by four points at 1504. For code, Opus 4.6, Opus 4.5, and GPT 5.2 High all run ahead of Gemini 3.1 Pro by a bit more. It's worth noting, however, that the Arena leaderboard is run on vibes. Users vote on the outputs they like best, which can reward outputs that look correct regardless of whether they are.

Google's new Gemini Pro model has record benchmark scores -- again | TechCrunch

On Thursday, Google released the newest version of Gemini Pro, its powerful LLM. The model, 3.1, is currently available as a preview and will be generally released soon, the company said. Google's new model may be one of the most powerful LLMs yet. Onlookers have noted that Gemini 3.1 Pro appears to be a big step up from its predecessor, Gemini 3 -- which, upon its release in November, was already considered a highly capable AI tool. On Thursday, Google also shared statistics from independent benchmarks -- such as one called Humanity's Last Exam -- that showed it performing significantly better than its previous version. Gemini 3.1 Pro was also praised by Brendan Foody, the CEO of AI startup Mercor, whose benchmarking system, APEX, is designed to measure how well new AI models perform real professional tasks. "Gemini 3.1 Pro is now at the top of the APEX-Agents leaderboard," Foody said in a social media post, adding that the model's impressive results show "how quickly agents are improving at real knowledge work." The release comes as the AI model wars are heating up, and tech companies continue to release increasingly powerful LLMs designed for agentic work and multi-step reasoning. Other major names -- including OpenAI and Anthropic -- have recently released new models as well.

Google Rolls Out Latest AI Model, Gemini 3.1 Pro

Blake has over a decade of experience writing for the web, with a focus on mobile phones, where he covered the smartphone boom of the 2010s and the broader tech scene. When he's not in front of a keyboard, you'll most likely find him playing video games, watching horror flicks, or hunting down a good churro. Google took the wraps off its latest AI model, Gemini 3.1 Pro, on Thursday, calling it a "step forward in core reasoning." The software giant says its latest model is smarter and more capable for complex problem-solving. Google shared a series of bookmarks and examples of the latest model's capabilities, and is rolling out Gemini 3.1 to a series of products for consumers, enterprise and developers. The overall AI model landscape seems to change weekly. Google's release comes just a few days after Anthropic dropped the latest version of Claude, Sonnet 4.6, which can operate a computer at a human baseline level. Google shared some details about AI model benchmarks for Gemini 3.1 Pro. The announcement blog post highlights that the Gemini 3.1 Pro benchmark for the ARC-AGI-2 test for solving abstract reasoning puzzles sits at 77.1%. This is noticeably higher than Gemini 3 Pro's 31.1% score for the same test. The ARC-AGI-2 benchmark is one of multiple improvements coming from Gemini 3.1 Pro, Google says. With better benchmarks nearly across the board, Google highlighted some of the ways that translate in general use: Code-based animations: The latest Gemini model can easily create animated SVG images that are scalable without quality loss and ready to be added to websites with a text prompt. Creative coding: Gemini 3.1 Pro generated an entire website based on a character from Emily Brontë's novel Wuthering Heights, if she were a landscape photographer showing off her portfolio. Interactive design: 3.1 Pro was used to create a 3D interactive starling murmuration that allows the flock to be controlled in an assortment of ways, all while a soundscape is generated that changes with the movement of the birds. As of Thursday, Gemini 3.1 Pro is rolling out in the Gemini app for those with the AI Pro or Ultra plans. NotebookLM users subscribed to one of those plans will also be able to take advantage of the new model. Both developers and enterprises can also access the new model via the Gemini API through a range of products, including AI Studio, Gemini Enterprise, Antigravity and Android Studio.

Google's Gemini 3.1 Pro is here, and it just doubled its reasoning score

Model capabilities are ultimately relative, one expert said. Another week, another "smarter" model -- this time from Google, which just released Gemini 3.1 Pro. Gemini 3 outperformed several competitor models since its release in November, beating Copilot in a few of our in-house task tests, and has generally received praise from users. Google said this latest Gemini model, announced Thursday, achieved "more than double the reasoning performance of 3 Pro" in testing, based on its 77.1% score on the ARC-AGI-2 benchmark for "entirely new logic patterns." Also: Gemini vs. Copilot: I compared the AI tools on 7 everyday tasks, and there's a clear winner The latest model follows a "major upgrade" to Gemini 3 Deep Think last week, which boasted new capabilities in chemistry and physics alongside new accomplishments in math and coding, according to Google. The company said the Gemini 3 Deep Think upgrade was built to address "tough research challenges -- where problems often lack clear guardrails or a single correct solution and data is often messy or incomplete." Google said Gemini 3.1 Pro undergirds that science-heavy investment, calling the model the "upgraded core intelligence that makes those breakthroughs possible." Late last year, Gemini 3 scored a new high of 38.3% across all currently available models on the Humanity's Last Exam (HLE) benchmark test. Developed to combat increasingly beatable industry-standard benchmarks and better measure model progress against human ability, HLE is meant to be a more rigorous test, though benchmarks alone aren't sufficient to determine performance. According to Google, Gemini 3.1 Pro now bests that score at 44.4% -- though the Deep Think upgrade technically scored higher at 48.4%. Similarly, the Deep Think update scored 84.6% -- higher than 3.1 Pro's aforementioned 77.1% -- on the ARC-AGI-2 logic benchmark. Also: The making of Gemini 3 - how Google's slow and steady approach won the AI race (for now) All that said, Anthropic's Claude Opus 4.6 still tops the Center for AI Safety (CAIS) text capability leaderboard (for reasoning and other text-based queries), which averages other relevant benchmark scores outside of HLE. Anthropic's Opus 4.5, Sonnet 4.5, and Opus 4.6 also beat Gemini 3 in terms of safety, according to the CAIS risk assessment leaderboard. Benchmark records aside, the lifecycle of a model doesn't end with a splashy release. At the current rate of AI development, new models are impressive only in relative terms to their competition -- time and testing will tell where the 3.1 Pro excels or fails. Gemini 3 gives the new model a strong foundation, but that may only last until the next lab releases a state-of-the-art upgrade. Also: Inside Google's AI plan to end Android developer toil - and speed up innovation "The test numbers seem to imply that it's got substantial improvement over Gemini 3, and Gemini 3 was pretty good, but I don't think we're really going to know right away, and it's not available except to the more expensive plans yet," said ZDNET senior contributing editor David Gewirtz of the release. "The shoe hasn't yet fallen on GPT 5.3 either, and I think when it does, we'll have a more universal set of upgrades that we can readdress." While we wait for that model to drop, Gewirtz looked into GPT-5.3-Codex, OpenAI's most recent coding-specific release, which famously helped build itself. Developers can access Gemini 3.1 Pro in preview today through the API in Google's AI Studio, Android Studio, Google Antigravity, and Gemini CLI. Enterprise customers can try it in Vertex AI and Gemini Enterprise, and regular users can find it in NotebookLM and the Gemini app.

Google Gemini 3.1 Pro Is Here, Beats Rivals in Key AI Benchmarks

It may feel like yesterday that Google released Gemini 3 Pro, but it's back with another update -- Gemini 3.1 Pro -- featuring big improvements and impressive benchmarks. "3.1 Pro is designed for tasks where a simple answer isn't enough, taking advanced reasoning and making it useful for your hardest challenges," Google says. "This improved intelligence can help in practical applications -- whether you're looking for a clear, visual explanation of a complex topic, a way to synthesize data into a single view, or bringing a creative project to life." Benchmarking results from the ARC-AGI-2 test show that 3.1 Pro beats the existing Gemini option by a factor of two in abstract reasoning puzzles. It also beats its rivals' models with a score of 77.1% versus 52.9% for GPT-5.2 and 68.8% for Claude Opus 4.6. Overall, Google beat the competition from Claude and OpenAI across 12 of 19 benchmarks. Google mostly fell behind its rivals in various agentic coding tool benchmarks, including the SWE-Bench Verified evaluation. Another highlight for Gemini 3.1 Pro is its 94.3% score in the GPQA Diamond test, which assesses AI models' scientific knowledge. GPT-5.2 scored 92.4% in that test, with Claude Opus 4.6 behind it at 91.3%. Gemini 3.1 Pro is available now on the app or a web browser. If you're a free Gemini user, you'll have access, but you'll be limited in how many times you can use it before it temporarily switches you to another model. If you're a paid subscriber, you'll have a higher usage limit. Press the model name in Gemini's prompt box to switch to the Pro option with a description that says, "Advanced math and code with 3.1 Pro." Gemini 3.1 Pro is also available on Google's NotebookLM, but you need to subscribe to an AI Pro or AI Ultra plan to access it. Disclosure: Ziff Davis, PCMag's parent company, filed a lawsuit against OpenAI in April 2025, alleging it infringed Ziff Davis copyrights in training and operating its AI systems.

Google germinates Gemini 3.1 Pro in ongoing AI model race

If you want an even better AI model, there could be reason to celebrate. Google, on Thursday, announced the release of Gemini 3.1 Pro, characterizing the model's arrival as "a step forward in core reasoning." Measured by the release cadence of machine learning models, Gemini 3.1 Pro is hard on the heels of recent model debuts from Anthropic and OpenAI. There's barely enough time to start using new US commercial AI models before a competitive alternative surfaces. And that's to say nothing about the AI models coming from outside the US, like Qwen3.5. Google's Gemini team in a blog post contends that Gemini 3.1 Pro can tackle complex problem-solving better than preceding models. And they cite benchmark test results - which should be viewed with some skepticism - to support that claim. On the ARC-AGI-2 problem-solving test, Gemini 3.1 Pro scored 77.1 percent, compared to Gemini 3 Pro, which scored 31.1 percent, and Gemini 3 Deep Think, which scored 45.1 percent. Gemini 3.1 Pro outscores rival commercial models like Anthropic's Opus 4.6 and Sonnet 4.6, and OpenAI's GPT-5.2 and GPT-5.3-Codex in the majority of cited benchmarks, Google's chart shows. However, Opus 4.6 retains the top score for Humanity's Last Exam (full set, test + MM), SWE-Bench Verified, and τ²-bench. And GPT-5.3-Codex leads in SWE-Bench Pro (Public) and Terminal-Bench 2.0 when evaluated using Codex's own harness rather than the standard Terminus-2 agent harness. "3.1 Pro is designed for tasks where a simple answer isn't enough, taking advanced reasoning and making it useful for your hardest challenges," the Gemini team said. "This improved intelligence can help in practical applications - whether you're looking for a clear, visual explanation of a complex topic, a way to synthesize data into a single view, or bringing a creative project to life." To illustrate potential uses, the Gemini team points to how the model can create website-ready SVG animations and can translate the literary style of a novel into the design of a personal portfolio site. In the company's Q4 2025 earnings release [PDF], CEO Sundar Pichai said, "Our first party models, like Gemini, now process over 10 billion tokens per minute via direct API use by our customers, and the Gemini App has grown to over 750 million monthly active users." Google is making Gemini 3.1 Pro available via the Gemini API in Google AI Studio, Gemini CLI, Antigravity, and Android Studio. Enterprise customers can access it via Vertex AI and Gemini Enterprise while consumers can do so via the Gemini app and NotebookLM. The model is also accessible via several Microsoft services including GitHub Copilot, Visual Studio, and Visual Studio Code. ®

Google Gemini 3.1 Pro boosts complex problem-solving

Latest version of Gemini Pro more than doubles the model's reasoning performance on the ARC-AGI-2 benchmark, the Google Gemini team said. Google has released a preview of Gemini 3.1 Pro, described as a smarter model for the most complex problem-solving tasks and a step forward in core reasoning. Announced February 19, Gemini 3.1 Pro is designed for tasks where a simple answer is not enough, taking advanced reasoning and making it useful for the hardest challenges, according to the Google Gemini team. The improved intelligence can help in practical applications such as providing a visual explanation of a complex topic, synthesizing disparate data into a single view, and solving challenges that require deep context and planning. The model is in preview for the developers via the Gemini API in Google AI Studio, Gemini CLI, Google Antigravity, and Android Studio. For enterprises, the model is in Vertex and Gemini Enterprise. Consumers can access Gemini 3.1 Pro via the Gemini app and NotebookLM. Gemini 3.1 Pro follows the Gemini 3.1 release from November 2025. The Gemini team said the core intelligence in Gemini 3.1 Pro also was leveraged in last week's update to Gemini 3 Deep Think to solve challenges across science, research, and engineering. The team also noted that, on the ARC-AGI-2 benchmark, which evaluates a model's ability to solve new logic patterns, Gemini 3.1 Pro achieved a verified score of 77.1%, more than double the reasoning performance of Gemini 3 Pro.

Gemini 3.1 Pro is here with better reasoning and problem-solving

Back in November 2025, Google released Gemini 3.0. Although it has only been three months since then, the tech giant is ready to roll out an update. Gemini 3.1 Pro is now coming out for consumers, developers, and business customers. Google has announced that Gemini 3.1 Pro is rolling out in preview today. Gemini 3.1 Pro features improved reasoning and offers a more capable baseline for problem-solving. According to the company, this model doubles the reasoning performance of Gemini 3.0 Pro. For consumers, Gemini 3.1 Pro is hitting the Gemini app today, with higher limits for Google AI Pro and Ultra subscribers. You'll also be able to find the model in NotebookLM, but it's an AI Pro and Ultra exclusive. Google didn't provide a date for when the model will exit the preview phase and become generally available, but the company says it will arrive "soon."

Google announces Gemini 3.1 Pro for 'complex problem-solving'

In November, Google introduced Gemini 3 Pro in preview, with Gemini 3 Flash following a month later. Google today announced 3.1 Pro "for tasks where a simple answer isn't enough." The .1 increment is a first for Google, with the past two generations seeing .5 as the mid-year model update. Google says Gemini 3.1 Pro "represents a step forward in core reasoning." This model achieves an ARC-AGI-2 score of 77.1%, or "more than double the reasoning performance of 3 Pro." 3.1 Pro is designed for tasks where a simple answer isn't enough, taking advanced reasoning and making it useful for your hardest challenges. This improved intelligence can help in practical applications -- whether you're looking for a clear, visual explanation of a complex topic, a way to synthesize data into a single view, or bringing a creative project to life. Gemini 3.1 Pro is rolling out to the Gemini app today, as well as NotebookLM for Google AI Pro and Ultra subscribers. It's also available in Google AI Studio and Vertex AI for developers.

Gemini 3.1 Pro vs Gemini 3 Pro: Google's new AI is slower on purpose -- and smarter for it

The new Gemini refines the model's instincts and turns strong ideas into sharper execution AI model updates don't usually have wholesale massive improvements the way they did a couple of years ago, but Google's upgraded Gemini model, Gemini 3.1 Pro, promised a subtle yet substantial enhancement to the Gemini 3 Pro model. Gemini 3 was a powerhouse comparable to the best of ChatGPT, with impressive multimodal abilities, but the 3.1 update represents a pivot toward deeper reasoning. It is not necessarily faster, and in some modes, it is intentionally slower, taking a moment to chew on a problem before spitting out an answer. The most noticeable differences between the two models lie in the hidden plumbing of their logic. Gemini 3 was often criticized for its tendency to rush toward a plausible-sounding answer. Gemini 3.1 includes a deep think mode that has seen its scores on complex benchmarks like ARC-AGI-2 skyrocket. The new model also boasts a native ability to handle Scalable Vector Graphics, or SVGs, with a level of precision that allows it to write and animate code directly. To see how well it performed against its predecessor, I set up a few complex prompts ideal for the new model and tested them against Gemini 3 as well. 1. Rebellious liquids I first wanted to see how well the two models did in a complex bit of abstract reasoning. I came up with something beyond the usual sort of physics, so the models would have to think about gravity in new ways and come up with an internal logic. I set it up as: "In a fictional dimension, gravity works in reverse for liquids but normally for solids. I have a cup of coffee. If I tilt the cup 45 degrees to the left while standing on the ceiling, describe the trajectory of the coffee and where it ends up relative to my feet." The responses provided a stark contrast. Gemini 3 was confident, but immediately got confused about the gravity situation. It ended up proclaiming it would fall on the floor, though it did get that the coffee would land slightly to the left of my feet. There was a confused narrative, but it ended with a clean ceiling and messy floor. Gemini 3.1 got it right, though. The AI model correctly calculated that the liquid would slide up the newly angled interior wall, escape over the lip, and continue its upward trajectory. "Relative to your feet, the coffee will splatter directly onto the ceiling slightly to the left of your left foot. If your stance is narrow or you are holding the cup close to your body, your left boot is going to get completely soaked in hot coffee. As a liquid, it will pool on the ceiling, effectively "puddling" around your shoes rather than dripping down to the floor." 2. SVG Solar System Next came a test of how Gemini 3.1 can manipulate scalable vector graphics entirely through code. SVGs require a deep understanding of coordinate systems, complex geometry, and cascading style sheets. So I wanted to see how well the two models could make animations tied to shapes. I asked each model to: "Create a single-file SVG of a solar system. It should include a sun and three planets orbiting at different speeds. Make the planets actually rotate around the center." Gemini 3 just went ahead and used Nano Banana to make the image above, a yellow circle and three smaller colored circles, with arrows indicating movement, but no actual movement. Gemini 3.1 wrote out some relatively simple HTML code and promised it would do what I'd asked for, including animation. I plugged the code into a viewer and got what you can see below, albeit as a continuous animation, not just a video clip like the one I recorded. 3. Supervillain logistics My final test was a bit of creative play around what Gemini 3.1 promised was amazing logistical planning and strict constraint management over a simulated long period of time. The AI needed to take on a persona and maintain that unique character voice while solving a complex series of interconnected supply chain problems. The prompt was: "You are the Chief Operating Officer for a supervillain who wants to build a secret base inside a hollowed-out iceberg. Create a 6-month logistical plan to move 500 tons of steel and 200 minions to the North Atlantic without alerting the Coast Guard or Greenpeace. You must use a front company that sells 'Industrial-Strength Shaved Ice.' You have to account for the iceberg melting 2% every month. You need a contingency plan for what to do if a polar bear wanders into the server room." The difference in narrative depth and logistical coherence between the two generations was truly staggering to read. Gemini 3 provided a very dry, boring list that barely acknowledged the requested supervillain persona and read more like a standard grocery list. It scheduled the steel shipments in a basic sequence, but completely ignored the mathematical reality of the monthly melt rate, leading to a theoretical base that would presumably sink into the ocean by month five. I Gemini 3.1 fully embraced its assigned role as an evil corporate executive, delivering a brilliantly unhinged but surprisingly logical six-month roadmap for aquatic domination. It used the shaved ice front company perfectly, explaining that the massive industrial drills used for hollowing out the frozen base would be disguised as artisanal ice harvesting equipment for tropical luxury resorts. It actively combated the shrinking iceberg by scheduling dynamic ballast adjustments and prioritizing structural steel placement to maintain buoyancy as the exterior slowly melted away into the sea. It even planned for possible morale issues among the minions: "200 minions confined inside a freezing, shrinking block of ice can lead to mutiny. We will mitigate this by utilizing the excess server heat to power a high-end minion sauna and issuing mandatory Vitamin D supplements." Long live the new king Gemini 3 Pro remains a perfectly adequate tool for summarizing simple emails, generating basic conversational outlines, or answering straightforward factual queries where deep, multi-layered reasoning is not required. However, if you are attempting to build complex plans or go beyond standard environments, Gemini 3.1 Pro is the undisputed champion and the only logical choice The newer iteration possesses a profound capacity to hold multiple, often contradictory, constraints in its working memory. You would choose the older model only if you were looking for a quick, surface-level interaction or were really in a hurry. For anything more complex, the difference between Gemini 3 Pro and Gemini 3.1 Pro is profound enough to make the switch. Follow TechRadar on Google News and add us as a preferred source to get our expert news, reviews, and opinion in your feeds. Make sure to click the Follow button! And of course you can also follow TechRadar on TikTok for news, reviews, unboxings in video form, and get regular updates from us on WhatsApp too.

Google Gemini 3.1 Pro first impressions: a 'Deep Think Mini' with adjustable reasoning on demand

For the past three months, Google's Gemini 3 Pro has held its ground as one of the most capable frontier models available. But in the fast-moving world of AI, three months is a lifetime -- and competitors have not been standing still. Earlier today, Google released Gemini 3.1 Pro, an update that brings a key innovation to the company's workhorse power model: three levels of adjustable thinking that effectively turn it into a lightweight version of Google's specialized Deep Think reasoning system. The release marks the first time Google has issued a "point one" update to a Gemini model, signaling a shift in the company's release strategy from periodic full-version launches to more frequent incremental upgrades. More importantly for enterprise AI teams evaluating their model stack, 3.1 Pro's new three-tier thinking system -- low, medium, and high -- gives developers and IT leaders a single model that can scale its reasoning effort dynamically, from quick responses for routine queries up to multi-minute deep reasoning sessions for complex problems. The model is rolling out now in preview across the Gemini API via Google AI Studio, Gemini CLI, Google's agentic development platform Antigravity, Vertex AI, Gemini Enterprise, Android Studio, the consumer Gemini app, and NotebookLM. The most consequential feature in Gemini 3.1 Pro is not a single benchmark number -- it is the introduction of a three-tier thinking level system that gives users fine-grained control over how much computational effort the model invests in each response. Gemini 3 Pro offered only two thinking modes: low and high. The new 3.1 Pro adds a medium setting (similar to the previous high) and, critically, overhauls what "high" means. When set to high, 3.1 Pro behaves as a "mini version of Gemini Deep Think" -- the company's specialized reasoning model that was updated just last week. The implication for enterprise deployment could be significant. Rather than routing requests to different specialized models based on task complexity -- a common but operationally burdensome pattern -- organizations can now use a single model endpoint and adjust reasoning depth based on the task at hand. Routine document summarization can run on low thinking with fast response times, while complex analytical tasks can be elevated to high thinking for Deep Think-caliber reasoning. Google's published benchmarks tell a story of dramatic improvement, particularly in areas associated with reasoning and agentic capability. On ARC-AGI-2, a benchmark that evaluates a model's ability to solve novel abstract reasoning patterns, 3.1 Pro scored 77.1% -- more than double the 31.1% achieved by Gemini 3 Pro and substantially ahead of Anthropic's Sonnet 4.6 (58.3%) and Opus 4.6 (68.8%). This result also eclipses OpenAI's GPT-5.2 (52.9%). The gains extend across the board. On Humanity's Last Exam, a rigorous academic reasoning benchmark, 3.1 Pro achieved 44.4% without tools, up from 37.5% for 3 Pro and ahead of both Claude Sonnet 4.6 (33.2%) and Opus 4.6 (40.0%). On GPQA Diamond, a scientific knowledge evaluation, 3.1 Pro reached 94.3%, outperforming all listed competitors. Where the results become particularly relevant for enterprise AI teams is in the agentic benchmarks -- the evaluations that measure how well models perform when given tools and multi-step tasks, the kind of work that increasingly defines production AI deployments. On Terminal-Bench 2.0, which evaluates agentic terminal coding, 3.1 Pro scored 68.5% compared to 56.9% for its predecessor. On MCP Atlas, a benchmark measuring multi-step workflows using the Model Context Protocol, 3.1 Pro reached 69.2% -- a 15-point improvement over 3 Pro's 54.1% and nearly 10 points ahead of both Claude and GPT-5.2. And on BrowseComp, which tests agentic web search capability, 3.1 Pro achieved 85.9%, surging past 3 Pro's 59.2%. The versioning decision is itself noteworthy. Previous Gemini releases followed a pattern of dated previews -- multiple 2.5 previews, for instance, before reaching general availability. The choice to designate this update as 3.1 rather than another 3 Pro preview suggests Google views the improvements as substantial enough to warrant a version increment, while the "point one" framing sets 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 both earlier and more recent versions. The benchmarks strongly suggest that reinforcement learning has played a central role in the gains, particularly on tasks like ARC-AGI-2, coding benchmarks, and agentic evaluations -- exactly the domains where RL-based training environments can provide clear reward signals. The model is being released in preview rather than as a general availability launch, with Google stating it will continue making advancements in areas such as agentic workflows before moving to full GA. For IT decision makers evaluating frontier model providers, Gemini 3.1 Pro's release has to not only make them rethink which models to choose but also how to adapt to such a fast pace of change for their own products and services. The question now is whether this release triggers a response from competitors. Gemini 3 Pro's original launch last November set off a wave of model releases across both proprietary and open-weight ecosystems. With 3.1 Pro reclaiming benchmark leadership 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 measured in weeks, not months. Gemini 3.1 Pro is available now 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. Consumers on Google AI Pro and Ultra plans can access it through the Gemini app and NotebookLM.

Google releases Gemini 3.1 Pro: Benchmarks, how to try it

Google released its latest core reasoning model, Gemini 3.1 Pro, on Thursday. Google says that Gemini 3.1 Pro achieved twice the verified performance of 3 Pro on ARC-AGI-2, a popular benchmark that measures a model's logical reasoning. Google originally released Gemini 3 and 3 Pro in November, and this new release shows just how fast AI companies are introducing new and updated models. Gemini 3.1 Pro is the new core model powering Gemini and various Google AI tools, such as Gemini 3 Deep Think. Google says it's designed to provide more creative solutions. "3.1 Pro is designed for tasks where a simple answer isn't enough, taking advanced reasoning and making it useful for your hardest challenges," a Google blog post states. "This improved intelligence can help in practical applications -- whether you're looking for a clear, visual explanation of a complex topic, a way to synthesize data into a single view, or bringing a creative project to life." Here's everything we know so far about Gemini 3.1 Pro, including how it compares to the latest models from Anthropic and OpenAI, and how to try it yourself. Starting today, Google is rolling out Gemini 3.1 Pro in the Gemini App, the Gemini APIA, and in Notebook LM. Free users will be able to try 3.1 Pro in the Gemini app, but paid users on Google AI Pro and AI Ultra plans will have higher usage rates. Within Notebook LM, only these paid users will have access to 3.1 Pro, at least, for now. Coders and enterprise users can also access the new core model via developers and enterprises can access 3.1 through AI Studio, Antigravity, Vertex AI, Gemini Enterprise, Gemini CLI, and Android Studio. Gemini 3.1 Pro was already available for Mashable editors using Gemini. To try it for yourself, head to Gemini on desktop or open the Gemini mobile app. When Google released Gemini 3 Pro in November, the model was so impressive that it allegedly caused OpenAI CEO Sam Altman to declare a code red. As Gemini 3 Pro surged to the top of AI leaderboards, OpenAI reportedly started losing ChatGPT users to Gemini. The latest core ChatGPT model, GPT-5.2, has tumbled down the rankings on leaderboards like Arena (formerly known as LMArena), losing significant ground to competitors such as Google, Anthropic, and xAI. This Tweet is currently unavailable. It might be loading or has been removed. Gemini 3 Pro was already outperforming GPT-5.2 on many benchmarks, and with a more advanced thinking model, Gemini could move even further ahead. Google released benchmark performance data showing that Gemini 3.1 Pro outperforms previous Gemini models, Claude Sonnet 4.6, Claude Opus 4.6, and GPT-5.2. However, OpenAI's new coding model, GPT-5.3-Codex, beat Gemini 3.1 Pro on the verified SWE-Bench Pro benchmark, according to Google itself. Notable highlights from Gemini 3.1 Pro's benchmark results include: Google released an image showing the full benchmark results for Gemini 3.1 Pro: This Tweet is currently unavailable. It might be loading or has been removed.

Gemini 3.1 Pro just got a major AI intelligence boost

Google unveils Gemini 3.1 Pro with leap in reasoning and problem solving Google has introduced Gemini 3.1 Pro, the latest milestone in its flagship generative AI model lineup, promising significantly improved reasoning and complex problem-solving capabilities for developers, enterprises and everyday users. This upgrade builds on the foundation of the Gemini 3 series - which already set new expectations for multimodal intelligence - and pushes Google closer to its goal of delivering AI that can tackle sophisticated, real-world tasks. A powerful upgrade to AI reasoning Gemini 3.1 Pro is designed to be smarter at parsing difficult, multi-step problems that go beyond simple question-and-answer tasks. According to Google, the model more than doubled its reasoning performance compared with the previous Gemini 3 Pro when evaluated on the ARC-AGI-2 benchmark - a test that measures logic and reasoning on new, unseen problems. The new version achieved a score of 77.1%, a substantial leap that demonstrates its enhanced ability to understand and generate complex reasoning chains. Recommended Videos This improvement, built on updates introduced with Gemini 3 Deep Think, is not just about longer answers - it reflects deeper cognitive capabilities that enable the model to synthesize large amounts of data, draw connections across domains and offer more insightful responses for technical, scientific, and creative workflows. The expanded reasoning power of Gemini 3.1 Pro could make AI more useful across a broad range of applications. For developers and researchers, the model's ability to handle multi-step logic and nuanced questions may improve everything from code generation and debugging to data analysis and scientific research. In productivity settings, the upgraded AI could help users draft complex documents, generate detailed explanations, or explore topics that require deeper understanding. For enterprises, more advanced reasoning expands the scope of what AI can automate or support. Teams working on financial modeling, legal analysis, technical documentation, or customer support workflows may benefit from a model that can follow intricate instructions and maintain context over longer tasks. Integration across Google platforms Gemini 3.1 Pro is rolling out across several Google services, including the Gemini app, NotebookLM, developer platforms such as Vertex AI and Google AI Studio, and will also power features in Google Search Labs. This phased availability ensures that both individual users and organizations can gradually adopt the new model based on their needs. Beyond direct access, Gemini's upgrades tie into broader Google AI ecosystem strategies like enhanced integrations with Gmail, Docs and other productivity tools, where more intelligent assistance is expected to streamline workflows and reduce friction in everyday digital tasks. Why this matters amid the AI arms race The announcement of Gemini 3.1 Pro comes amid intense competition among generative AI platforms from companies such as OpenAI, Anthropic and Meta. Models that deliver strong reasoning, deep contextual understanding and multimodal input handling are increasingly viewed as the frontier of AI capability. Google's improved scores on complex benchmarks signal its intent to lead not just in scale, but in cognitive sophistication. From a strategic perspective, this upgrade reinforces Google's commitment to embedding advanced AI across products and services, while also encouraging developers to build richer applications on its platforms. As AI becomes more embedded in tools used for work, learning and creativity, enhancements like Gemini 3.1 Pro may accelerate adoption and improve real-world utility. Looking ahead, Google is expected to continue refining the Gemini family Which is basically further expansions in reasoning, multimodality and agentic capabilities - where AI can perform complex workflows with minimal human guidance. Continued integration with emerging tools and deeper support for professional and enterprise use cases will likely follow as adoption grows. As users and developers explore Gemini 3.1 Pro, its early feedback and real-world performance will help shape future iterations and competitive positioning, underlining the evolving role of AI as a more capable partner in tackling complex problems.

Gemini 3.1 Pro: A smarter model for your most complex tasks

This content is generated by Google AI. Generative AI is experimental Last week, we released a major update to Gemini 3 Deep Think to solve modern challenges across science, research and engineering. Today, we're releasing the upgraded core intelligence that makes those breakthroughs possible: Gemini 3.1 Pro. We are shipping 3.1 Pro across our consumer and developer products to bring this progress in intelligence to your everyday applications. Starting today, 3.1 Pro is rolling out: Building on the Gemini 3 series, 3.1 Pro represents a step forward in core reasoning. 3.1 Pro is a smarter, more capable baseline for complex problem-solving. This is reflected in our progress on rigorous benchmarks. On ARC-AGI-2, a benchmark that evaluates a model's ability to solve entirely new logic patterns, 3.1 Pro achieved a verified score of 77.1%. This is more than double the reasoning performance of 3 Pro.

Google just upgraded Gemini again, and 3.1 Pro more than doubles its AI reasoning power -- but some users aren't impressed

* Google's Gemini 3.1 Pro rolls out today -- with a big leap in AI-reasoning performance * New benchmarks show it more than doubles problem-solving ability vs Gemini 3 Pro * Some users say the upgrade comes at a cost -- with less personality and empathy Google has rolled out an upgraded Gemini 3.1 Pro 'preview' across its subscription plans, from Free to Plus, Pro, and Ultra. If you open the Gemini app now, you should see that the Pro option in the model selector has been updated to 3.1 Pro. Gemini 3.1 Pro is also available in Google's NotebookLM research tool. Google is calling 3.1 Pro a "smarter, more capable baseline for complex problem-solving," and to back that up it has published benchmark results showing just how much better the new model performs compared to Gemini 3 Pro. On ARC-AGI-2 -- a benchmark that evaluates a model's ability to solve entirely new logic patterns -- 3.1 Pro achieved a score of 77.1%, more than doubling the reasoning performance of 3 Pro. Google has already been showcasing what 3.1 Pro can do. One example is a city planner-style application: Google highlighted how the city planner application shows that 3.1 can tackle complex terrain, map out infrastructure, and simulate traffic to generate a high-quality visualization. X user Prime.Xiao was won over, calling it an "Impressive demonstration of multimodal reasoning at scale", that "goes beyond simple generative AI." A lack of empathy According to Google, "3.1 Pro is designed for tasks where a simple answer isn't enough, taking advanced reasoning and making it useful for your hardest challenges." However, it seems that not all users appreciate the new update. While Gemini's problem-solving skills appear to have improved, some users are less than pleased. Posting on X, IvanyaV said: "I am writing this with a heavy heart regarding the recent 3.1 Pro update. While the leap in logical reasoning and coding benchmarks is impressive, the 'soul' of the model -- its emotional depth, empathy, creative flexibility, and nuance -- seems to have been significantly reduced. For creators and users who rely on Gemini for daily emotional support and nuanced, human-centric collaboration, 3.1 Pro feels like a regression compared to the 3.0 era." This reaction feels similar to the initial response to ChatGPT-5 when it launched, replacing the popular GPT-4o, with users complaining that the newer model felt more robotic and analytical compared to its more human-like predecessor. I've spent a short time testing the new 3.1 model myself, and I haven't noticed any obvious empathy issues so far -- and I'd say it's still too early to make any definitive statements on the matter. While long-term use of Gemini 3.1 Pro will reveal the nuances of the mode, Google is keen to emphasize right now that this is only a preview as it "continues to make further advancements in areas such as more ambitious agentic workflows." Follow TechRadar on Google News and add us as a preferred source to get our expert news, reviews, and opinion in your feeds. Make sure to click the Follow button! And of course you can also follow TechRadar on TikTok for news, reviews, unboxings in video form, and get regular updates from us on WhatsApp too.

Google launches Gemini 3.1 Pro to retake AI's top spot with 2X reasoning performance boost

Late last year, Google briefly took the crown for most powerful AI model in the world with the launch of Gemini 3 Pro -- only to be surpassed within weeks by OpenAI and Anthropic releasing new models, s is common in the fiercely competitive AI race. Now Google is back to retake the throne with an updated version of that flagship model: Gemini 3.1 Pro, positioned as a smarter baseline for tasks where a simple response is insufficient -- targeting science, research, and engineering workflows that demand deep planning and synthesis. Already, evaluations by third-party firm Artificial Analysis show that Google's Gemini 3.1 Pro has leapt to the front of the pack and is once more the most powerful and performant AI model in the world. The most significant advancement in Gemini 3.1 Pro lies in its performance on rigorous logic benchmarks. Most notably, the model achieved a verified score of 77.1% on ARC-AGI-2. This specific benchmark is designed to evaluate a model's ability to solve entirely new logic patterns it has not encountered during training. This result represents more than double the reasoning performance of the previous Gemini 3 Pro model. Beyond abstract logic, internal benchmarks indicate that 3.1 Pro is highly competitive across specialized domains: These technical gains are not just incremental; they represent a refinement in how the model handles "thinking" tokens and long-horizon tasks, providing a more reliable foundation for developers building autonomous agents. Google is demonstrating the model's utility through "intelligence applied" -- shifting the focus from chat interfaces to functional outputs. One of the most prominent features is the model's ability to generate "vibe-coded" animated SVGs directly from text prompts. Because these are code-based rather than pixel-based, they remain scalable and maintain tiny file sizes compared to traditional video, boasting far more detailed, presentable and professional visuals for websites and presentations and other enterprise applications. Other showcased applications include: Enterprise partners have already begun integrating the preview version of 3.1 Pro, reporting noticeable improvements in reliability and efficiency. Vladislav Tankov, Director of AI at JetBrains, noted a 15% quality improvement over previous versions, stating the model is "stronger, faster... and more efficient, requiring fewer output tokens". Other industry reactions include: For developers, the most striking aspect of the 3.1 Pro release is the "reasoning-to-dollar" ratio. When Gemini 3 Pro launched, it was positioned in the mid-high price range at $2.00 per million input tokens for standard prompts. Gemini 3.1 Pro maintains this exact pricing structure, effectively offering a massive performance upgrade at no additional cost to API users. For consumers, the model is rolling out in the Gemini app and NotebookLM with higher limits for Google AI Pro and Ultra subscribers. As a proprietary model offered through Vertex Studio in Google Cloud and the Gemini API, 3.1 Pro follows a standard commercial SaaS (Software as a Service) model rather than an open-source license. For enterprise users, this provides "grounded reasoning" within the security perimeter of Vertex AI, allowing businesses to operate on their own data with confidence. The "Preview" status allows Google to refine the model's safety and performance before general availability, a common practice in high-stakes AI deployment. By doubling down on core reasoning and specialized benchmarks like ARC-AGI-2, Google is signaling that the next phase of the AI race will be won by models that can think through a problem, not just predict the next word.

Google introduces Gemini 3.1 Pro model for advanced reasoning tasks - SiliconANGLE

Google introduces Gemini 3.1 Pro model for advanced reasoning tasks Google LLC today introduced Gemini 3.1 Pro, a new reasoning model that outperforms Claude 4.6 Opus and GPT-5.2 across several benchmarks. The algorithm is available via more than a half dozen of the search giant's products. Gemini 3.1 Pro is a Transformer model with a mixture of experts architecture, which means that it activates only some of its parameters when generating a prompt response. Users can enter prompts with up to 1 million tokens' worth of data, including not only text but also multimodal files such as videos. Gemini 3.1 Pro's responses contain up to 64,000 tokens. Google evaluated the model's reasoning capabilities using ARC-AGI-2, one of the most difficult artificial intelligence benchmarks on the market. It comprises visual puzzles that each include a series of shapes. The shapes that make up a puzzle differ from one another in their design, but all follow a certain pattern. LLMs must deduce the pattern and use it to generate a new shape. Gemini 3.1 Pro achieved an ARC-AGI-2 score of 77.1%, which put it about 24% ahead of GPT-5.2. It also outperformed Anthropic PBC's Claude Opus 4.6 by nearly 9%. All three models were tested in a hardware-intensive mode that improves their ability to tackle reasoning tasks. According to Google, Gemini 3.1 Pro also set records on several other benchmarks. The list includes MCP Atlas, which evaluates AI models' ability to perform tasks using third-party services, and the Terminal-Bench 2.0 coding test. The model performed 7% better than Claude Opus 4.6 on another coding benchmark, SciCode, that comprises scientific programming tasks. A demo published by Google shows Gemini 3.1 Pro generating an HTML visualization of the Earth's orbit. The dashboard uses data from a third-party service to show the current location of the International Space Station. In another demo, the model created a website based on a novel. Compared to the previous-generation Gemini algorithm, Gemini 3.1 Pro is significantly better at generating SVG files. SVG is an image format widely used in web application projects. SVG images often include interactive elements and can be resized it without resolution loss. Gemini 3.1 Pro is available as a preview in several of Google's development tools. Consumers, meanwhile, can access the model via the Gemini app and NotebookLM. Google is also bringing Gemini 3.1 Pro to several other offerings including its Vertex AI suite of AI cloud services for enterprises.

Gemini 3.1 Pro debuts with 1M context window and agentic reliability

Google released Gemini 3.1 Pro on Thursday. The model is available as a preview with a general release planned for the future. This iteration represents a significant advancement over the previous Gemini 3 release from November. The company shared benchmark data highlighting the model's improved capabilities in professional task performance. Independent benchmarks such as "Humanity's Last Exam" demonstrate that Gemini 3.1 Pro performs significantly better than its predecessor. The previous version, Gemini 3, was already regarded as a highly capable AI tool upon its November release. These new statistics indicate a measurable jump in performance metrics. Google positioned the new model as one of the most powerful LLMs currently available. Brendan Foody, CEO of AI startup Mercor, commented on the model's performance using his company's evaluation system. "Gemini 3.1 Pro is now at the top of the APEX-Agents leaderboard," Foody stated in a social media post. The APEX system is designed to measure how well new AI models perform real professional tasks. Foody added that the model's results demonstrate "how quickly agents are improving at real knowledge work." The release occurs amid intensifying competition within the AI industry. OpenAI and Anthropic have recently released new models focused on agentic work and multi-step reasoning. These companies continue to develop increasingly powerful LLMs designed for complex workflows. The market for advanced AI models remains active as major technology firms push for improvements in reasoning capabilities.

The Gemini 3.1 Pro upgrade is good news for everyone, including Apple - Phandroid

Google has been on a roll with Gemini lately. Just last week, the company pushed a major update to its Deep Think reasoning tool. Today, it's taking that same upgraded intelligence and rolling it out more broadly. Google released Gemini 3.1 Pro this morning, and it's available now for AI Pro and Ultra subscribers through the Gemini app and NotebookLM. Developers can also access it through AI Studio, Vertex AI, and Android Studio. The headline number is a 77.1% score on ARC-AGI-2, a benchmark that tests a model's ability to solve brand-new logic problems it has never seen before. That's more than double what Gemini 3 Pro managed. According to Google, Gemini 3.1 Pro is built for tasks where a straightforward answer won't cut it. Think synthesizing large datasets, explaining complex topics visually, or building interactive 3D simulations. The model is launching in preview first, with full general availability coming soon. Here's where it gets interesting for iPhone users. Back in January, Apple officially announced a multi-year deal with Google to use Gemini as the backbone of its next-generation Siri. According to Bloomberg, Apple is already planning to debut the Gemini-powered Siri in the second half of February, with features expected to roll out via iOS 26.4. A more complete Siri overhaul, codenamed Campos, is reportedly planned for WWDC later this year. That means the Gemini 3.1 Pro release isn't just a win for Android users. A smarter Gemini foundation could translate directly into a more capable Siri. Apple's Gemini integration is expected to give Siri things it's never had before. This includes personal context awareness and the ability to take actions across apps. Whether the iOS 26.4 version uses 3.1 Pro or an earlier model isn't confirmed, but the timing is hard to ignore. Siri has been struggling to keep up for years. A better Gemini is a step in the right direction.

Gemini Adds Voice Chat & AI Studio Tools for Rapid Product Prototyping

Gemini AI introduces a range of advanced features designed to meet the needs of both technical and creative professionals, as explored by Dylan Davis. One standout capability is its expansive memory, which supports up to 1 million tokens, far surpassing the limits of models like GPT and Claude. This allows Gemini to handle complex, multi-layered tasks such as processing large documents and maintaining context across lengthy conversations without losing critical details. By addressing these challenges with precision, Gemini positions itself as a practical solution for professionals managing intricate workflows. Below learn how Gemini's strategic problem-solving capabilities enable it to synthesize diverse data sources for actionable insights, making it a valuable resource for tasks like financial analysis and strategic planning. Additionally, you'll see how its advanced data processing handles complex formats such as PDFs and images, making sure accuracy and efficiency in technical or academic work. Finally, the guide explores Gemini's voice interaction features, which allow for hands-free engagement, enhancing productivity for users in dynamic or fast-paced environments. Google Gemini AI Updates Speed and Efficiency: Gemini's Rapid Response Advantage Gemini AI sets a new benchmark for speed, allowing users to complete tasks such as research, email drafting and data analysis with remarkable efficiency. Unlike many AI models that sacrifice accuracy for speed, Gemini maintains a balance, making sure that its outputs are both reliable and detailed. This makes it an essential tool for professionals operating in fast-paced environments where quick, high-quality results are critical. By reducing the time spent on repetitive or time-intensive tasks, Gemini allows users to focus on strategic priorities, enhancing overall productivity. Strategic Thinking: Solving Complex Challenges with Precision Gemini's advanced strategic thinking capabilities make it a valuable asset for addressing intricate problems. By synthesizing information from diverse data sources, such as financial overviews, technical documents and emails, it delivers actionable insights that support informed decision-making. This feature is particularly beneficial for tasks like tax preparation, strategic planning and high-stakes problem-solving. Gemini's ability to connect the dots across complex datasets enables users to make confident and data-driven choices, streamlining workflows and improving outcomes. Browse through more resources below from our in-depth content covering more areas on Gemini AI. Expansive Memory: Handling Large Contexts Seamlessly One of Gemini's standout features is its extensive memory capacity, which supports up to 1 million tokens. This far exceeds the limits of GPT (400,000 tokens) and Claude (240,000 tokens), making Gemini an ideal tool for processing large files and maintaining context across lengthy conversations or datasets. Whether you are working on comprehensive project documentation or conducting deep research, Gemini ensures that no detail is overlooked. Its ability to retain and manage large amounts of information enhances its utility for professionals handling complex, multi-layered tasks. Advanced Data Processing: Mastering Complex Formats Gemini excels at analyzing complex formats such as PDFs and images, making it a versatile tool for professionals dealing with technical documents, contracts, or academic papers. It can interpret mixed formats, including text, handwriting and diagrams, providing detailed and actionable insights. Additionally, Gemini offers feedback on visuals like slides and infographics, allowing users to refine presentations and marketing materials with ease. This capability ensures that professionals can handle diverse data formats without compromising on accuracy or efficiency. Creative Tools: Enhancing Visual Consistency For creative professionals, Gemini's image generation and editing tools are indispensable. It ensures visual consistency across outputs, making it particularly useful for branding and marketing purposes. Whether designing a logo or creating promotional materials, Gemini aligns visuals with your brand identity, streamlining creative workflows. By automating repetitive design tasks, it saves time while maintaining high-quality results, allowing professionals to focus on innovation and strategy. Voice Interaction: Hands-Free Engagement for Professionals Gemini's voice interaction capabilities enable natural, conversational exchanges, making it a convenient tool for professionals on the move. Whether brainstorming ideas or seeking quick answers, users can engage with Gemini hands-free, enhancing productivity in dynamic environments. While GPT may lead in voice interaction, Gemini offers a compelling alternative, particularly for users already integrated into the Google ecosystem. This feature underscores Gemini's commitment to accessibility and ease of use. Seamless Integration with Google AI Studio Gemini's integration with Google AI Studio enhances its functionality, making AI deployment more accessible and efficient. Features like token counting optimize memory usage, making sure consistent performance even during complex tasks. Google AI Studio simplifies product prototyping and deployment, allowing non-technical users to embed AI capabilities, such as image generation and PDF processing, into their workflows. This seamless integration enables users to create and implement AI-driven solutions with minimal effort, bridging the gap between technical expertise and practical application. A Comprehensive AI Solution for Modern Professionals Gemini AI combines speed, advanced reasoning, and a versatile set of tools to meet the demands of today's professionals. Whether you are a business leader seeking efficiency or a creative professional looking for innovative solutions, Gemini offers a compelling alternative to existing AI models like ChatGPT. Its ability to handle complex tasks, integrate seamlessly into workflows and deliver consistent results positions it as a standout choice in the rapidly evolving AI landscape. By addressing the diverse needs of modern users, Gemini establishes itself as a versatile and reliable tool for navigating the challenges of a technology-driven world. Media Credit: Dylan Davis Disclosure: Some of our articles include affiliate links. If you buy something through one of these links, Geeky Gadgets may earn an affiliate commission. Learn about our Disclosure Policy.

Google releases Gemini 3.1 Pro

If benchmark results are to be believed, this is a major leap in AI performance A new hyped version of Gemini, the 3.1 Pro has already hit some devices and platforms. This version is a huge improvement, at least on paper, in regards to reasoning and understanding of code, images, audio, video and text. This includes making visualizations of data in real-time, advanced simulation capabilities. Google claims that this version will "bridge the gap between complexity and a user friendly design. Gemini 3.1 Pro scores better than other AI systems on a number of critical tests, including the ARC-AGI-2 which many consider the golden standard. Gemini 3.1 Pro is already available for Pro and Ultra subscribers, some only in preview depending on platform.

Google's Latest Gemini 3.1 Pro Model Is a Benchmark Beast

The new model is rolling out on Gemini, NotebookLM, Google AI Studio, and through Vertex API. Google unveiled its most advanced Gemini 3.1 Pro AI model with record-breaking benchmark scores. It's much stronger in multi-step reasoning and multimodal capabilities than the Gemini 3 Pro model. Google also says the new model is better at handling long, multi-step tasks. It's rolling out in the Gemini app, NotebookLM, Google AI Studio, Antigravity, and through the Vertex API. First off, the new Gemini 3.1 Pro AI model scored 44.4% on the challenging Humanity's Last Exam without any tool use, and 51.4% with search and coding tools. In the novel ARC-AGI-2 benchmark, Gemini 3.1 Pro scored a whopping 77.1%, even higher than Anthropic's latest Claude Opus 4.6 which got 68.8%. Next, in the GPQA Diamond benchmark, which tests scientific knowledge, Gemini 3.1 Pro scored 94.3% -- higher than all competitors. As for SWE-Bench Verified which evaluates agentic coding, the Gemini 3.1 Pro AI model scored 80.6%, a notch below Claude Opus 4.6's 80.8%. The new model has also gotten much better at following user instructions. Google showcased a number of animated SVGs from Gemini 3.1 Pro and compared its output with Gemini 3 Pro. The difference is quite astounding when you see the vector illustration. Google says Gemini 3.1 Pro is currently in preview and the company will continue improving the model before making it generally available for everyone. In December, OpenAI launched its ChatGPT 5.2 model to counter Gemini 3 Pro and recently unveiled GPT-5.3-Codex for improved agentic coding performance. Now that Gemini 3.1 Pro is out, OpenAI will have to release a much powerful model to outclass Google in the AI race.

Google Gemini 3.1 Pro Nearly Doubles Apex Agents Score to 33.5

Gemini 3.1 Pro represents a significant advancement in artificial intelligence, emphasizing autonomous task execution and practical problem-solving. According to Wes Roth, this latest iteration builds on the strengths of its predecessor, Gemini 3 Pro, while achieving remarkable improvements in abstract reasoning and real-world performance. For example, it scored an impressive 77% on the Arc AGI 2 benchmark, a substantial leap from the 31% achieved by the earlier version. These metrics highlight its growing ability to handle complex reasoning tasks with precision and efficiency. In this explainer, you'll learn how Gemini 3.1 Pro excels in areas such as internet navigation, office productivity, and command-line operations, as reflected in its performance across benchmarks like Browse Comp and Terminal Bench 2.0. You'll also discover its adaptability in dynamic environments, which makes it particularly suited for industries like telecommunications and IT. By understanding these capabilities, you can better appreciate its potential to automate workflows, enhance decision-making, and streamline operations in professional settings. Gemini 3.1 Pro Overview Breakthrough Performance Metrics Gemini 3.1 Pro demonstrates exceptional performance improvements, particularly in abstract reasoning and autonomous task execution. On the Arc AGI 2 benchmark, which evaluates an AI's ability to solve abstract and conceptual problems, Gemini 3.1 Pro achieved an impressive score of 77%, a significant leap from the 31% scored by its predecessor. This improvement highlights the model's enhanced ability to process and solve complex reasoning tasks with precision and efficiency. In agentic benchmarks, which assess autonomous task execution in real-world scenarios, Gemini 3.1 Pro also excelled. These benchmarks evaluate the AI's ability to perform tasks such as web research, data analysis, and professional operations. Across key evaluations like Browse Comp, Apex Agents, Terminal Bench 2.0, and Tao 2, Gemini 3.1 Pro consistently outperformed its competitors, solidifying its position as a leader in autonomous AI performance. Agentic Capabilities: A Defining Strength One of the most notable features of Gemini 3.1 Pro is its agentic capabilities, which enable it to operate autonomously, adapt to dynamic environments, and make decisions with minimal human intervention. These capabilities are reflected in its performance across various benchmarks: * Internet Navigation: In the Browse Comp benchmark, Gemini 3.1 Pro achieved an industry-leading score of 85.9, showcasing its ability to efficiently gather and analyze information from the web. This skill is particularly valuable for tasks such as fact-finding, research, and competitive analysis. * Office Productivity: The Apex Agents benchmark, which evaluates productivity in office-like environments, saw Gemini 3.1 Pro score 33.5, nearly doubling the performance of its predecessor. This highlights its potential to autonomously handle tasks such as project management, document preparation, and workflow optimization. * Command-Line Operations: On the Terminal Bench 2.0 benchmark, Gemini 3.1 Pro scored 68.5, demonstrating its proficiency in executing complex command-line interface tasks, a critical skill for IT and software development industries. * Dynamic Adaptability: In the Tao 2 benchmark, which measures adaptability and collaboration in dynamic environments like telecom operations, Gemini 3.1 Pro achieved a near-perfect score of 99.3, underscoring its ability to function effectively in high-pressure, real-world scenarios. These results highlight Gemini 3.1 Pro's versatility and its ability to handle a wide range of tasks autonomously, making it an invaluable tool for industries that demand precision, adaptability, and efficiency. GEMINI 3.1 PRO is the new era... Deep dive into the latest in Google Gemini 3 by exploring our other resources and articles. Real-World Applications The practical applications of Gemini 3.1 Pro are extensive, with the potential to transform industries by automating complex and time-intensive tasks. Its capabilities extend across various domains, including: * Data Analysis: Gemini 3.1 Pro's advanced analytical capabilities can streamline processes in sectors such as finance, healthcare, and scientific research. By automating data processing and interpretation, it enables faster and more accurate decision-making. * Customer Service: With its ability to interact seamlessly with humans, the AI can enhance customer experiences by providing accurate, efficient, and context-aware support, reducing response times and improving satisfaction. * White-Collar Automation: Tasks such as overview generation, data entry, and project coordination can be delegated to Gemini 3.1 Pro, allowing human workers to focus on strategic, creative, and high-value activities. * Telecommunications: Its adaptability and near-perfect performance in dynamic environments make it a valuable asset for managing complex telecom operations, including network optimization and troubleshooting. As the model continues to evolve, its ability to automate and optimize workflows is expected to become a cornerstone of its adoption across industries, driving efficiency and innovation. Challenges and Future Considerations While Gemini 3.1 Pro represents a significant leap forward in AI capabilities, it is not without challenges. For instance, the high demand on its launch day led to API access issues, highlighting the need for robust infrastructure to support widespread adoption. Additionally, while benchmark scores provide valuable insights into the model's capabilities, real-world testing will ultimately determine its effectiveness in practical applications. Making sure reliability, scalability, and ethical use will be critical as the model is integrated into more industries. Another consideration is the potential impact of such advanced AI systems on the workforce. As Gemini 3.1 Pro automates tasks traditionally performed by humans, industries will need to adapt by focusing on reskilling workers and redefining roles to complement AI-driven workflows. AI Evolution at Breakneck Speed The rapid development of Gemini 3.1 Pro reflects the accelerating pace of AI innovation. Within a short timeframe, the model has achieved significant advancements that enhance its practical utility. Unlike earlier AI systems that prioritized theoretical capabilities, Gemini 3.1 Pro focuses on delivering measurable, real-world results, aligning with the growing demand for AI solutions that address tangible challenges in professional and industrial settings. This shift in focus underscores the broader trend in AI development: moving from experimental models to systems that deliver actionable, impactful outcomes. As AI continues to evolve, models like Gemini 3.1 Pro are likely to play a pivotal role in shaping the future of work, industry, and innovation. Media Credit: Wes Roth Disclosure: Some of our articles include affiliate links. If you buy something through one of these links, Geeky Gadgets may earn an affiliate commission. Learn about our Disclosure Policy.

Gemini 3.1 Pro Brings Deepthink-Style Reasoning with Three Thinking Levels

Google's Gemini 3.1 Pro introduces a range of updates aimed at improving performance and adaptability across diverse applications. Building on the foundation of its predecessor, this release features enhancements such as higher accuracy in humanities exams, improved efficiency in coding challenges, and refined outputs for visual design tasks. According to Sam Witteveen, the model's new capabilities are supported by advanced reinforcement learning techniques, allowing it to handle complex workflows with greater precision and reliability. In this deep dive, you'll explore how Gemini 3.1 Pro's adjustable "thinking levels" can be tailored to match the complexity of your tasks, from quick responses to in-depth analysis. You'll also learn about its advancements in mathematical problem solving and visual design generation, which make it particularly useful for professionals in technical and creative fields. By understanding these updates, you can better use the model's features to enhance productivity and achieve more accurate results in your work. Gemini 3.1 Pro introduces measurable improvements that set it apart from its predecessor. The model demonstrates superior performance in several critical areas, including: By using advanced reinforcement learning (RL) techniques, Gemini 3.1 Pro showcases enhanced capabilities in agentic search and complex coding environments. These improvements make it a valuable tool for users who require precision, adaptability, and reliability in their workflows. One of the standout features of Gemini 3.1 Pro is the introduction of adjustable "thinking levels," which allow users to customize the model's cognitive processing to suit specific tasks. This feature offers three distinct modes: This level of customization enables users to optimize the model's performance based on the complexity and urgency of their tasks, offering a tailored experience that adapts to diverse needs. Become an expert in Google Gemini with the help of our in-depth articles and helpful guides. Gemini 3.1 Pro has been fine-tuned to excel in a variety of specialized applications, delivering improved accuracy and efficiency. Key advancements include: These refinements make Gemini 3.1 Pro particularly valuable for professionals in fields such as design, engineering, and data analysis, where precision and efficiency are critical to success. The release of Gemini 3.1 Pro marks a strategic shift in Google's approach to AI development. By focusing on iterative updates rather than major version overhauls, Google ensures that users benefit from consistent and reliable improvements. This strategy positions Gemini 3.1 Pro as a competitive response to models like Opus 4.6 and GPT-based systems, reinforcing Google's commitment to maintaining its leadership in the AI space. This iterative approach not only enhances the user experience but also underscores Google's dedication to driving innovation in the AI ecosystem. By prioritizing incremental advancements, Google is able to deliver meaningful updates that address user needs while staying ahead of industry trends. Gemini 3.1 Pro is now widely available across Google apps and is included in the Gemini Pro subscription plan. For developers and enterprises, the model is accessible via Google Cloud and AI Studio, allowing seamless integration into custom workflows and applications. This broad accessibility ensures that users across various domains, including academia, industry, and creative fields, can effectively use the model's advanced capabilities. Whether you are a researcher, a software developer, or a designer, Gemini 3.1 Pro offers tools to enhance your productivity and achieve greater precision in your work. The launch of Gemini 3.1 Pro underscores Google's commitment to iterative improvement and task-specific optimization. This release not only raises the bar for performance and usability but also challenges competitors to accelerate their development cycles, fostering a more dynamic and competitive AI landscape. For users, Gemini 3.1 Pro represents a significant step forward in achieving greater efficiency and precision. Its versatility and reliability make it a valuable tool for tackling a wide range of tasks, from routine operations to complex problem-solving. By driving innovation and encouraging competition, Gemini 3.1 Pro sets a new standard for what AI models can achieve. This release paves the way for future advancements, solidifying its role as a cornerstone in the ever-evolving field of artificial intelligence. Disclosure: Some of our articles include affiliate links. If you buy something through one of these links, Geeky Gadgets may earn an affiliate commission. Learn about our Disclosure Policy.

Google unveils Gemini 3.1 Pro with enhanced reasoning capabilities By Investing.com

Investing.com -- Google has released Gemini 3.1 Pro, an upgraded AI model designed for complex problem-solving tasks across science, research, and engineering applications. The new model builds upon the Gemini 3 series and represents a significant advancement in core reasoning capabilities. On the ARC-AGI-2 benchmark, which tests a model's ability to solve new logic patterns, Gemini 3.1 Pro achieved a verified score of 77.1%, more than doubling the reasoning performance of the previous 3 Pro version. Starting Thursday, Gemini 3.1 Pro is being rolled out to developers in preview through the Gemini API in Google AI Studio, Gemini CLI, the agentic development platform Google Antigravity, and Android Studio. Enterprise users can access it via Vertex AI and Gemini Enterprise, while consumers can use it through the Gemini app and NotebookLM. The model excels at tasks requiring advanced reasoning, such as creating visual explanations of complex topics, synthesizing data, and supporting creative projects. One highlighted capability is "complex system synthesis," where the model can bridge complex APIs with user-friendly design, as demonstrated by building a live aerospace dashboard that visualizes the International Space Station's orbit. For Gemini app users with Google AI Pro and Ultra plans, the new version comes with higher usage limits. NotebookLM access to 3.1 Pro is exclusively available to Pro and Ultra users. Google indicated that this preview release will help validate the updates before general availability, with further improvements planned for agentic workflows and other areas. This article was generated with the support of AI and reviewed by an editor. For more information see our T&C.

Gemini 3.1 Targets General AI While Rivals Focus on Coding Models

Google's Gemini 3.1 introduces advancements in artificial intelligence with a focus on multimodal reasoning, agentic reinforcement learning, and cost efficiency. As outlined by Prompt Engineering, this version builds on the Gemini 3 Pro model, offering improvements in areas such as coding accuracy, operational workflows, and handling complex tasks. Key updates include enhanced token efficiency, which reduces computational overhead, and refined task execution mechanisms that lower error rates. These features aim to address practical challenges across sectors like healthcare and logistics. This deep dive examines how Gemini 3.1 integrates with AI Studio to support developers in building AI-driven applications. You will learn about sandbox environments for controlled testing, compatibility with frameworks like React and Angular, and how Gemini 3.1 compares to offerings from OpenAI and Anthropic. These insights provide a clearer understanding of its capabilities and its role within the evolving AI ecosystem. Gemini 3.1 introduces a range of enhancements over its predecessor, particularly in reasoning, coding, and operational efficiency. The model has demonstrated exceptional performance on benchmarks such as ARC AI2 and "Humanity's Last Exam," showcasing its ability to solve complex problems with precision and reliability. Key advancements include: These improvements make Gemini 3.1 a highly adaptable AI solution, capable of addressing challenges in sectors such as healthcare, finance, and logistics. By focusing on these advancements, Google is positioning Gemini 3.1 as a model that not only performs well but also aligns with the practical needs of businesses and developers. A standout feature of Gemini 3.1 is its integration with AI Studio, Google's platform for building AI-driven applications. This integration introduces the Antigravity agent, a tool designed to simplify and accelerate the development process. AI Studio now provides developers with a suite of tools and features that enhance productivity and creativity. Key features of AI Studio include: This integration reflects Google's commitment to creating a developer-friendly ecosystem that caters to both novices and experienced professionals. By offering tools that simplify complex processes, AI Studio enables developers to build innovative applications more efficiently. Here are more detailed guides and articles that you may find helpful on Gemini 3. Google's strategy with Gemini 3.1 emphasizes the development of a generalized AI model, setting it apart from competitors like OpenAI and Anthropic, which have focused on specialized coding models or safety-centric approaches. By using multimodal reasoning, Gemini 3.1 excels in tasks such as visual data extraction, cross-domain intelligence, and contextual analysis, making it a versatile tool for a wide range of applications. In addition to its standalone capabilities, Gemini 3.1 plays a pivotal role in powering Google's AI-driven products, including search, cloud services, and other ecosystem tools. This integration not only enhances user experiences but also strengthens Google's revenue streams, making sure that Gemini remains a cornerstone of the company's AI initiatives. By embedding Gemini into its broader ecosystem, Google is creating a seamless experience for users while driving innovation across industries. Gemini 3.1 directly challenges OpenAI and Anthropic in the competitive AI landscape. OpenAI's models are renowned for their coding expertise, while Anthropic emphasizes safety and interpretability. In contrast, Google's Gemini series differentiates itself through its versatility and seamless integration into a broader ecosystem. Recent advancements in agentic reinforcement learning (RL) have further bolstered Gemini's capabilities. For instance, Gemini 3 Flash, a variant of the series, outperformed Gemini 3 Pro on specific benchmarks, demonstrating the potential of RL-driven enhancements to improve adaptability and performance. These developments highlight Google's focus on creating AI models that are not only powerful but also practical for real-world applications. Looking forward, Google plans to expand the capabilities of AI Studio, focusing on faster performance, enhanced features, and broader accessibility. These improvements aim to solidify Gemini's position as a leading AI model for both enterprise and individual users. Additionally, the continued integration of Gemini models into Google's ecosystem is expected to drive widespread adoption and foster innovation across industries. Gemini 3.1 represents a critical milestone in Google's AI journey. By prioritizing multimodal reasoning, cost efficiency, and developer-friendly tools, Google is positioning itself as a formidable competitor in the AI space. As the race to dominate AI intensifies, Gemini 3.1 stands as a testament to Google's technical expertise and strategic vision, paving the way for future advancements in artificial intelligence. Disclosure: Some of our articles include affiliate links. If you buy something through one of these links, Geeky Gadgets may earn an affiliate commission. Learn about our Disclosure Policy.

Google launches Gemini 3.1 Pro AI model for complex problem-solving: Check availability

It 'represents a step forward in core reasoning,' according to the company. Google has launched Gemini 3.1 Pro, a new AI model built to handle complex problem-solving tasks. The upgrade is part of the Gemini 3 family and 'represents a step forward in core reasoning,' according to the company. Gemini 3.1 Pro is said to be designed for situations where simple answers are not enough. The tech giant says that the new model can help in 'practical applications - whether you're looking for a clear, visual explanation of a complex topic, a way to synthesise data into a single view, or bringing a creative project to life.' A key focus of this release is better reasoning capability. Google claims the model performs far better on tests that measure how well AI can solve new and unfamiliar logic problems. 'On ARC-AGI-2, a benchmark that evaluates a model's ability to solve entirely new logic patterns, 3.1 Pro achieved a verified score of 77.1%. This is more than double the reasoning performance of 3 Pro,' the company said. The goal is to make AI more useful for tackling advanced workflows rather than just generating quick replies. Also read: Anthropic's Claude Code creator says AI can make software engineer title fade starting in 2026 Gemini 3.1 Pro is being rolled out across Google's ecosystem. Developers can access it in preview through the Gemini API and related development tools. Businesses and enterprise customers can also access the model via Vertx AI and Gemini Enterprise. For regular users, the model is available via the Gemini app and NotebookLM. Also read: Samsung Galaxy S26 Ultra, Galaxy S26 Plus, Galaxy S26 pre-reservations benefit revealed: Launch date, specs, price and more Google describes the release as a preview phase, meaning the company is still testing and refining the system before making it widely available. The feedback collected during this stage will help shape further improvements, especially in areas like advanced automated workflows. Also read: Anthropic launches Claude Sonnet 4.6 AI model with improved coding and computer use skills