PaLM 2

blog.google uses cookies from Google to deliver and enhance the quality of its services and to analyze traffic. Learn moreUnderstood Introducing PaLM 2 May 10, 2023 · 5 min read Share x.com Facebook LinkedIn Mail Copy link Z Zoubin Ghahramani Vice President, Google DeepMind Share x.com Facebook LinkedIn Mail Copy link Updated September 12, 2023When you look back at the biggest breakthroughs in AI over the last decade, Google has been at the forefront of so many of them. Our groundbreaking work in foundation models has become the bedrock for the industry and the AI-powered products that billions of people use daily. As we continue to responsibly advance these technologies, there’s great potential for transformational uses in areas as far-reaching as healthcare and human creativity.Over the past decade of developing AI, we’ve learned that so much is possible as you scale up neural networks — in fact, we’ve already seen surprising and delightful capabilities emerge from larger sized models. But we’ve learned through our research that it’s not as simple as “bigger is better,” and that research creativity is key to building great models. More recent advances in how we architect and train models have taught us how to unlock multimodality, the importance of having human feedback in the loop, and how to build models more efficiently than ever. These are powerful building blocks as we continue to advance the state of the art in AI while building models that can bring real benefit to people in their daily lives. Introducing PaLM 2 Building on this work, today we’re introducing PaLM 2, our next generation language model. PaLM 2 is a state-of-the-art language model with improved multilingual, reasoning and coding capabilities.Multilinguality: PaLM 2 is more heavily trained on multilingual text, spanning more than 100 languages. This has significantly improved its ability to understand, generate and translate nuanced text — including idioms, poems and riddles — across a wide variety of languages, a hard problem to solve. PaLM 2 also passes advanced language proficiency exams at the “mastery” level.Reasoning: PaLM 2’s wide-ranging dataset includes scientific papers and web pages that contain mathematical expressions. As a result, it demonstrates improved capabilities in logic, common sense reasoning, and mathematics.Coding: PaLM 2 was pre-trained on a large quantity of publicly available source code datasets. This means that it excels at popular programming languages like Python and JavaScript, but can also generate specialized code in languages like Prolog, Fortran and Verilog.As we continue to develop PaLM 2, we’ll also be introducing version updates for PaLM 2 as it is implemented into products. A versatile family of models Even as PaLM 2 is more capable, it’s also faster and more efficient than previous models — and it comes in a variety of sizes, which makes it easy to deploy for a wide range of use cases. We’ll be making PaLM 2 available in four sizes from smallest to largest: Gecko, Otter, Bison and Unicorn. Gecko is so lightweight that it can work on mobile devices and is fast enough for great interactive applications on-device, even when offline. This versatility means PaLM 2 can be fine-tuned to support entire classes of products in more ways, to help more people. Powering over 25 Google products and features At I/O today, we announced over 25 new products and features powered by PaLM 2. That means that PaLM 2 is bringing the latest in advanced AI capabilities directly into our products and to people — including consumers, developers, and enterprises of all sizes around the world. Here are some examples:PaLM 2’s improved multilingual capabilities are allowing us to expand Bard to new languages, starting today. Plus, it’s powering our recently announced coding update.Workspace features to help you write in Gmail and Google Docs, and help you organize in Google Sheets are all tapping into the capabilities of PaLM 2 at a speed that helps people get work done better, and faster.Med-PaLM 2, trained by our health research teams with medical knowledge, can answer questions and summarize insights from a variety of dense medical texts. It achieves state-of-the-art results in medical competency, and was the first large language model to perform at “expert” level on U.S. Medical Licensing Exam-style questions. We're now adding multimodal capabilities to synthesize information like x-rays and mammograms to one day improve patient outcomes. Med-PaLM 2 will open up to a small group of Cloud customers for feedback later this summer to identify safe, helpful use cases. Sorry, your browser doesn't support embedded videos, but don't worry, you can download it and watch it with your favorite video player! Sec-PaLM is a specialized version of PaLM 2 trained on security use cases, and a potential leap for cybersecurity analysis. Available through Google Cloud, it uses AI to help analyze and explain the behavior of potentially malicious scripts, and better detect which scripts are actually threats to people and organizations in unprecedented time.Since March, we've been previewing the PaLM API with a small group of developers. Starting today, developers can sign up to use the PaLM 2 model, or customers can use the model in Vertex AI with enterprise-grade privacy, security and governance. PaLM 2 is also powering Duet AI for Google Cloud, a generative AI collaborator designed to help users learn, build and operate faster than ever before. Advancing the future of AI PaLM 2 shows us the impact of highly capable models of various sizes and speeds — and that versatile AI models reap real benefits for everyone. Yet just as we’re committed to releasing the most helpful and responsible AI tools today, we’re also working to create the best foundation models yet for Google.Our Brain and DeepMind research teams have achieved many defining moments in AI over the last decade, and we’re bringing together these two world-class teams into a single unit, to continue to accelerate our progress. Google DeepMind, backed by the computational resources of Google, will not only bring incredible new capabilities to the products you use every day, but responsibly pave the way for the next generation of AI models.We’re already at work on Gemini — our next model created from the ground up to be multimodal, highly efficient at tool and API integrations, and built to enable future innovations, like memory and planning. Gemini is still in training, but it’s already exhibiting multimodal capabilities never before seen in prior models. Once fine-tuned and rigorously tested for safety, Gemini will be available at various sizes and capabilities, just like PaLM 2, to ensure it can be deployed across different products, applications, and devices for everyone’s benefit. POSTED IN: Related stories AI Products 3 Google AI tools that helped me get my new job By Abby Crowe Feb 26, 2026 Developer tools New developer tools for Google AI Pro and Ultra subscribers By Niv Govindaraju & Bala Muthukrishnan Jan 27, 2026 AI Products The latest AI news we announced in December By Keyword Team Dec 29, 2025 AI Products Google's year in review: 8 areas with research breakthroughs in 2025 By Jeff Dean & Demis Hassabis & James Manyika Dec 23, 2025 AI Products 60 of our biggest AI announcements in 2025 By Zahra Thompson Dec 22, 2025 AI Products 40 of our most helpful AI tips from 2025 By Molly McHugh-Johnson Dec 19, 2025 . Jump to position 1 Jump to position 2 Jump to position 3 Jump to position 4 Jump to position 5 Jump to position 6 Let’s stay in touch. Get the latest news from Google in your inbox. Subscribe No thanks Survey Help us improve The Keyword with a one-question surveyYes NoThis survey is anonymous. All responses will be aggregated and used only for analysis to improve our services. Did this article provide the level of detail you were looking for? Yes, I got what I needed No, I wanted more technical depth No, I wanted a simpler overview I was looking for something else entirely ✅Thank you! --- blog.google uses cookies from Google to deliver and enhance the quality of its services and to analyze traffic. Learn moreUnderstood About the Keyword The Keyword is Google's official publication for the latest company news, updates and behind-the-scenes storytelling. Our contributors include hundreds of the world’s leading engineers, product managers and internet pioneers who work directly on products and services like Google Search, Ads, Gemini, Maps, Android, Chrome, Pixel and Play.The Keyword also regularly features opinions and perspectives from Google’s senior leadership, including CEO Sundar Pichai, President & Chief Investment Officer Ruth Porat and President of Global Affairs Kent Walker.Launched in 2016, the Keyword replaced more than a dozen blogs covering all aspects of Google’s technology, impact and culture. Since then, the Keyword has published thousands of articles and hundreds of opinion pieces covering topics ranging from our work in generative AI with Google DeepMind to the latest Android updates. The Keyword is also the central hub for breaking news and announcements from annual events like Google I/O and Made by Google.The Keyword is designed to grow and change as Google does, so as we introduce new products and technologies they will have a home here. And alongside the biggest news of the day, we’ll also share quick product tips and ideas to make your life easier.Where you’re from is an important part of who you are, and the name “the Keyword” is a nod to the origins of search. Keywords also unlock information, which is what we hope to do here — provide a central, official source of news about who we are, what we do and what we value. (Plus, we like that it has “word” in it.)In short: If there’s anything you want to know about what’s happening at Google right now, the Keyword is where you should start. Noteworthy authorsGooglers from across the company contribute to the Keyword — ranging from engineers and designers to doctors and scientists. Here are a few examples: Sundar PichaiSundar is the CEO of Alphabet and Google and serves on Alphabet’s Board of Directors. Under his leadership, Google has been focused on developing products and services, powered by the latest advances in AI, that offer help in moments big and small. Demis HassabisDemis Hassabis is the co-founder and CEO of Google DeepMind, one of the world’s leading AI research groups. DeepMind has produced landmark research breakthroughs such as AlphaGo, the first program to beat the world champion at the game of Go, and AlphaFold, which was heralded as a solution to the 50-year grand challenge of protein folding. James ManyikaJames is a senior vice president at Google. He focuses on technology and society, including areas ranging from AI to the future of work and the digital economy, computing infrastructure, sustainability, and other areas that have the potential for broad impact on society. Prior to Google, he spent over 25 years at McKinsey & Company, including serving on its board, and is now senior partner emeritus and chair emeritus of the McKinsey Global Institute (MGI). Ruth PoratAs President and Chief Investment Officer, Ruth is responsible for the company’s corporate investments and investment vehicles, including GV and CapG, the Other Bets investment portfolio, Real Estate and Workplace Services, and other company infrastructure. She was previously SVP and CFO of Alphabet and Google. Kent WalkerKent is the President of Global Affairs at Google and Alphabet, where he works at the intersection of technology, policy and legal issues. He joined Google in 2006, and currently oversees the company’s engagement with governments around the world, content policies, and philanthropic efforts. Rick OsterlohRick Osterloh leads Platforms & Devices, a group of products and services that bring the best of Google’s innovation in AI, software and hardware to users, partners, and developers. This includes the overall Android platform, Google Play, Chrome browser, ChromeOS, Google One, Google Photos, the Pixel portfolio, Nest, Fitbit and more. He works closely with partners across the ecosystem, and oversees multiple Google products that serve more than a billion users each. How to follow usYou can also find the latest news and stories from Google on our official social channels — including Instagram, YouTube, LinkedIn, X and TikTok — and our weekly email newsletter. Get more stories from Google in your inbox. Get more stories from Google in your inbox. Email address Your information will be used in accordance with Google's privacy policy. Subscribe Done. Just one step more. Check your inbox to confirm your subscription. You are already subscribed to our newsletter. You can also subscribe with a different email address . Contact usNeed assistance or want to report a problem with a Google product or service? Here’s the best place to start.If you are a member of the press, contact press@google.com.Google Headquarters1600 Amphitheatre ParkwayMountain View, CA 94043650-253-0000 Let’s stay in touch. Get the latest news from Google in your inbox. Subscribe No thanks --- blog.google uses cookies from Google to deliver and enhance the quality of its services and to analyze traffic. Learn moreUnderstood Google DeepMind News from Google DeepMind about how we're building AI responsibly to benefit everyone. Google DeepMind Create new worlds in Project Genie with these 4 tips Learn more about Google DeepMind’s Project Genie and how to write prompts to create your own worlds. Collection Introducing Gemini 3 Gemini 3, our most intelligent model, combines all of Gemini’s capabilities together so you can bring any idea to life. View the collection Gemini models A new era of intelligence with Gemini 3 By Sundar Pichai Gemini App Gemini 3 brings upgraded smarts and new capabilities to the Gemini app By Josh Woodward Search Google Search with Gemini 3: Our most intelligent search yet By Elizabeth Reid All the Latest Measuring progress toward AGI: A cognitive frameworkBroadening advanced AI education across AfricaPlatform 37 and The AI Exchange: new spaces for AI innovation and discoveryThe latest AI news we announced in FebruaryCreate new worlds in Project Genie with these 4 tipsAsk a Techspert: What’s a world model?Load more stories Let’s stay in touch. Get the latest news from Google in your inbox. Subscribe No thanks --- blog.google uses cookies from Google to deliver and enhance the quality of its services and to analyze traffic. Learn moreUnderstood Google Research Google Research Groundsource: using AI to help communities better predict natural disasters Groundsource is a new AI-powered methodology from Google Research that transforms millions of public records into actionable data. All the Latest Quantum frontiers may be closer than they appearBuilding superconducting and neutral atom quantum computersOur new study explores how AI can reduce the climate impact of air travel.Broadening advanced AI education across AfricaGroundsource: using AI to help communities better predict natural disastersHonoring the art of the possible this International Women’s DayLoad more stories Let’s stay in touch. Get the latest news from Google in your inbox. Subscribe No thanks