Artificial Intelligence ## NVIDIA Powers the World’s AI. And Yours. Upgrade to advanced AI with RTX AI PCs and accelerate gaming, creating, productivity, and development. [Shop Now](https://www.nvidia.com/en-gb/ai-on-rtx/#shop) Blog ## Perplexity Brings Local AI Agents to NVIDIA DGX Spark Perplexity Portable Computer lets users run a private personal AI agent locally on NVIDIA DGX Spark, without token limits. Users can switch between local and cloud models while connecting the agent to everyday tools. [Learn More](https://blogs.nvidia.com/blog/local-ai-open-source-models-agents-nemotron/#spark-perplexity) Blog ## NVIDIA Accelerates Google DeepMind's DiffusionGemma for Local AI The new DiffusionGemma open model generates text in parallel and is optimized for NVIDIA RTX PRO, DGX Spark, and GeForce RTX GPUs. [Learn More](https://blogs.nvidia.com/blog/rtx-ai-garage-local-gemma-diffusion) Blog ## NVIDIA Accelerates AI Agents on RTX PCs and DGX Spark NVIDIA OpenShell is coming to Windows, alongside new agentic AI optimizations and creator app upgrades—including NVIDIA Broadcast 2.2, plus upcoming RTX acceleration for Adobe apps and DLSS Ray Reconstruction for Blender Cycles. [Learn More](https://blogs.nvidia.com/blog/rtx-ai-garage-computex-spark-local-agents) Previous Next 1. Games 2. Agentic AI 3. RTX AI Garage blog 4. Energy **Introducing the NVIDIA RTX Spark Superchip** The fusion of NVIDIA AI and RTX graphics in a single chip redefines Windows PCs and delivers amazing creating, AI development, and gaming. [Learn More](https://www.nvidia.com/en-gb/products/rtx-spark.md) ## Future Ready. Present Deals. Game and create with GeForce RTX 50 Series, powered by AI. [Shop All](https://www.nvidia.com/en-gb/geforce/campaigns/holiday-deals.md) \*Terms and Conditions Apply When it comes to AI PCs, the best have NVIDIA GeForce RTX™ GPUs inside. That’s because the same technology powering world-leading AI innovation is built into every RTX GPU, giving you the power to do the extraordinary. From class to work to entertainment, with RTX-powered AI, you’re getting the most advanced AI experiences available on Windows PCs. ![NVIDIA GeForce RTX AI PC Overview Video](https://www.nvidia.com/content/dam/en-zz/Solutions/ai-on-rtx/nvidia-rtx-ai-pc-video-ari.jpg) Consent for Optional Cookies (googleCookiePolicyLink)YouTube sets performance, advertising, and other optional cookies(/googleCookiePolicyLink) when you watch embedded videos. To watch this video, you need to turn on optional cookies for the site. By clicking “Accept and Play Video,” you will automatically turn on advertising and other optional cookies for the site and accept our (nvidiaTermsOfServiceLink)Terms of Service(/nvidiaTermsOfServiceLink) (which contains important waivers). Please see our (nvidiaPrivacyPolicyLink)Privacy Policy(/nvidiaPrivacyPolicyLink) and (nvidiaCookiePolicyLink)Cookie Policy(/nvidiaCookiePolicyLink) for more information. Cancel Accept and Play Video Alternatively, you can (youtubeLink)watch this video on YouTube(/youtubeLink). ### Advanced AI PCs Get the fastest performance in over 950 AI-accelerated apps and games. ### AI-Enhanced Experiences The sharpest streaming, fastest editing, and improved video conferencing. ### Best in Class Conquer your coursework with [AI-accelerated STEM apps](https://www.nvidia.com/en-gb/geforce/laptops/stem-majors.md) for students. ### Private On-Device Run AI locally while your data stays secure on your device—no internet required. ## Multiply Your Performance ## Develop **4x** ### Faster AI Model Fine-Tuning ## Game **12x** ### Higher FPS in Modern Games ## Create **10x** ### Faster Image Generation ## Chat **4x** ### Faster LLM Text Generation As compared to a laptop without a GeForce RTX Laptop GPU. Hardware: RTX 5060 Laptop GPU (100W TGP), vs. M4 Pro (Develop & Create) or Ultra 9 285H with Arc 140T (Game & Chat) Develop: Avg samples/sec transformer fine tuning Pytorch BERT BS=16, mixed-precision enabled Game: Avg across Avowed, Hogwarts Legacy, Marvel Rivals, S.T.A.L.K.E.R. 2 at 1080p, highest settings with DLSS 4. Create: Time to generate 20 images with Stable Diffusion in ComfyUI. Chat: Avg tokens/sec interference LLAMA 3.1 8B, ISL/OSL 100/100, quantized int4. ## Use NIM Microservices in Your Favorite Apps Experience the power of NIM with the tools you already know and love. [[NVIDIA ChatRTX](https://www.nvidia.com/en-gb/ai-on-rtx/chatrtx.md)](https://www.nvidia.com/en-gb/ai-on-rtx/chatrtx/) [[Microsoft AI Toolkit](https://aka.ms/aitoolkit)](https://aka.ms/aitoolkit) [[AnythingLLM](https://docs.anythingllm.com/nvidia-nims/introduction)](https://docs.anythingllm.com/nvidia-nims/introduction) [[CrewAI](https://docs.crewai.com/concepts/llms#local-nvidia-nim-deployed-using-wsl2)](https://docs.crewai.com/concepts/llms#local-nvidia-nim-deployed-using-wsl2) [[ComfyUI](https://www.comfy.org/)](https://www.comfy.org/) [[Langflow](https://docs.langflow.org/integrations-nvidia-ingest-wsl2)](https://docs.langflow.org/integrations-nvidia-ingest-wsl2) [[Langchain](https://www.langchain.com/)](https://www.langchain.com/) [[FlowiseAI](https://docs.flowiseai.com/integrations/langchain/chat-models/nvidia-nim)](https://flowiseai.com/) To develop with NIM microservices on RTX AI PCs, start your journey [here](https://build.nvidia.com/search/all?q=Run-on-RTX). 1. NIM Microservices 2. AI Blueprints ## NIM Microservices NVIDIA NIM microservices deliver cutting-edge AI models optimized for RTX GPUs. With one click, you can access tools for language, vision, speech, design, and more. Use them in apps like ChatRTX, LM Studio, and ComfyUI to streamline workflows and power AI solutions. It’s AI, simplified and supercharged—ready when you are. ### Language Models Discover advanced language models, like [Meta AI’s Llama family](http://llama.com), built for reasoning, coding, and multilingual tasks. ### Speech Models Harness the power of speech models like the NVIDIA Riva family of models and [NVIDIA Maxine Studio Voice](https://build.nvidia.com/nvidia/studiovoice) to transcribe audio to text, generate natural-sounding speech, and more. ### Retrieval Models Easily ingest, store, and search through a variety of data—text, images, and tables—using powerful retrieval models like the [NVIDIA Retrieval QA Llama3.2 1b Embedding Model](https://build.nvidia.com/nvidia/llama-3_2-nv-embedqa-1b-v2). ### Visual Design Models Leverage powerful visual design models like [FLUX.1-dev](https://build.nvidia.com/black-forest-labs/flux_1-dev) to create high-quality images from text prompts. ### Vision Models Tap into powerful vision models like [NV-CLIP](https://build.nvidia.com/nvidia/nvclip), designed to interpret and analyze images, text, and videos. Previous Next 1. Tile0 2. Tile1 3. Tile2 4. Tile3 5. Tile4 6. Tile5 ## AI Blueprints Experience NVIDIA AI Blueprints on your RTX AI PC—pre-built workflows for tasks like turning PDFs into podcasts and generating images. Then, take control to modify, build, and create your own AI workflows. ### 3D Object Generation Accelerate your creative pipeline. Describe your vision and instantly generate a curated set of ready-to-use 3D assets. Refresh, refine, and convert previews into fully editable models for Blender—no waiting, no roadblocks. Prototype entire scenes in minutes with pre-built, NIM-powered workflows designed to keep up with your imagination. [Get Started](https://build.nvidia.com/nvidia/object-generation-3d) [Watch the Video](https://www.youtube.com/watch?v=bi146-Z7KvI) ### 3D-Guided Generative AI With state-of-the-art FLUX models, you can fully control your AI image generation process via a 3D rendering app like Blender. Simply move objects around and adjust the camera to generate images of the scene. No more complex prompts—just intuitive, fast, and powerful image generation. [Get Started](https://build.nvidia.com/nvidia/genai-3d-guided) [Watch the Video](https://youtu.be/ITlTTG9qyY8?si=C0iiD3Bokb7_aMgJ) ### PDF to Podcast Transform any PDF into an AI-generated podcast so you can learn on the go—no reading required. This blueprint brings your PDFs to life, from tables to text, transforming them into dynamic audio experiences. Dive deeper with interactive conversations, allowing you to explore any topic in more detail, anytime, anywhere. [Get Notified](https://developer.nvidia.com/ai-apps-for-rtx-pcs/rtx-nim-notify-me/) AI for You ## Do It All in the Blink of an AI AI is the answer to all the times you wish you had an assistant. Unlock a new era of productivity with NVIDIA-powered AI doing the heavy lifting for you. Agentic AI ## Local AI Agents on RTX: Private, Secure, and Free Step into the next phase of the AI revolution with autonomous agents like OpenClaw that can help with everyday tasks by working with the files and context on your PC. Powered by RTX-accelerated AI running locally, you can keep agents running around the clock at no extra cost, while your data stays safely on your device. ### Get Started With OpenClaw on NVIDIA RTX GPUs and DGX Spark Build and run powerful AI agents locally with OpenClaw on NVIDIA RTX GPUs and DGX Spark™. Follow our [step-by-step guide](https://www.nvidia.com/en-gb/geforce/news/open-claw-rtx-gpu-dgx-spark-guide.md) to get started with fast, RTX-accelerated performance. ### Control and Optimize Your System With G-Assist [NVIDIA Project G-Assist](https://www.nvidia.com/en-us/software/nvidia-app/g-assist.md) is an AI assistant that helps you tune, control, and optimize your system. Powered by a local RTX-accelerated AI model, it understands what you’re doing on your PC, and recommends intelligent optimizations to help you get the most out of your hardware. ### AI for Productivity and Entertainment ### AI: Your Everyday Edge From your favorite apps to your favorite shows, RTX-powered AI takes your everyday activities and makes them smarter, faster, sharper, and more enjoyable. With AI on your side, there’s nothing holding you back. ### Faster, Smarter AI Stay ahead with the latest AI models the moment they drop—running faster, smoother, fully private on your RTX-powered PC. Add your files for smarter, more personalized results. Experience it now through your favorite tools like LM Studio, AnythingLLM, and Nexa Hyperlink. ### Transform Your Room Into a Home Studio Level up your video conferencing calls, voice chats, and live streams with the NVIDIA Broadcast app. Offering AI-powered noise removal, background replacement, and more. ### Upgrade Your Video Viewing RTX Video transforms internet video into crystal clear 4K HDR video with AI super resolution and video HDR. Available in Chrome, Edge, and Firefox browsers and VLC. ### AI for Creators ### Imagination Made Real Supercharge your content creation with over [100 AI-enabled and RTX-accelerated creative apps](https://www.nvidia.com/en-gb/studio/software.md)—including the Adobe Creative Cloud suite, Blender, DaVinci Resolve, OBS Studio, and the exclusive [NVIDIA Studio](https://www.nvidia.com/en-gb/studio.md) suite of AI tools. ### High-Speed Image Generation Generate images and videos in Stable Diffusion [faster on NVIDIA RTX GPUs](https://www.youtube.com/watch?v=5KO4BHJSAZk), accelerated by NVIDIA TensorRT™. ### AI-Enhanced Video Editing Supercharge your video editing process with AI effects and tools in DaVinci Resolve, Adobe Premiere Pro, and more. ### AI-Powered 3D Design Enjoy real-time viewport rendering, upscaling, accelerated rendering, and more in leading 3D creative apps. ### AI for Gamers ### Gaming Powered by AI Level up your gaming experience with RTX and AI-powered technology enhancing everything from frame rates to fidelity. Plus, enjoy portable, powerful laptops with AI-powered Max-Q technologies to optimize your laptop’s performance. ### Supreme Speed. Superior Visuals. Powered by AI. NVIDIA DLSS is a revolutionary suite of neural rendering technologies that uses AI to boost FPS, reduce latency, and improve image quality. ### Remaster the Classics With RTX With [NVIDIA RTX Remix](https://www.nvidia.com/en-gb/geforce/rtx-remix.md) you can easily capture game assets, enhance materials with gen AI, and create stunning RTX remasters with full ray tracing and DLSS. ### Powerful, Portable, and Quiet Supercharge your laptop with the advanced [Max-Q](https://www.nvidia.com/en-gb/geforce/laptops/max-q-technologies.md) suite of AI-powered technologies that optimize your performance, power, and acoustics for peak efficiency. ### AI for Developers ### Easy, Private, and On-Device From running your own chatbot on your PC to training custom models and integrating AI into games, developing and running AI apps is easier and faster with NVIDIA tools and RTX GPUs. Plus, your data stays on your PC—no uploads required. ### Develop and Deploy AI on RTX Experiment, develop, and optimize with the latest AI technologies on RTX AI PCs. [Learn More](https://developer.nvidia.com/ai-apps-for-rtx-pcs) ### Accelerate AI on RTX GPUs Discover how to get maximum performance across popular inference backends, including Windows ML, Ollama, and PyTorch. [Learn More](https://developer.nvidia.com/ai-apps-for-rtx-pcs/inference-backends) ### Add AI Features to Your App Jumpstart your development with curated and pre-optimized AI SDKs and models for NVIDIA RTX GPUs. [Learn More](https://developer.nvidia.com/ai-apps-for-rtx-pcs/sdks-models) **Performance** ## Oh Chip—That’s Fast The AI processors in every GeForce RTX GPU deliver chart-busting levels of performance across the most demanding games, apps, and workflows. 1. ### Laptop 2. ### Desktop **Relative Performance (min)** Tested on GeForce RTX 5090 Laptop GPU with Intel Core Ultra 9 285H CPU, Apple M4 Max, and Intel Core Ultra 9 285H CPU with Intel Arc 140T Graphics | Gaming: Average FPS across Avowed, Hogwarts Legacy, Marvel Rivals, and S.T.A.L.K.E.R. 2: Heart of Chornobyl at 1440p with highest settings. DLSS: Super Resolution, Multi-Frame Generation on. Laptop without RTX GPU not capable of these features | Content Creation: Average performance across AI image generation (Stable Diffusion, ComfyUI), 3D rendering (Autodesk Arnold), photo editing (ON1 AI resize), video editing (DaVinci Resolve) | Productivity: Average tokens/second for LLM inference (Llama 3.1 8B, ISL/OSL = 100/100, quantized to int4) Development: Average samples/second for transformer fine tuning using PyTorch with BERT, BS=16. Mixed-precision enabled. **Relative Performance (FPS)** Tested on GeForce RTX 5090 Desktop GPU with Intel® Core™ i9-14900K CPU, and Apple Mac Studio (M3 Ultra). Content Creation: Average performance across AI image generation (Flux, ComfyUI), 3D rendering (VRAY 6.0 benchmark), photo editing (ON1 AI resize), video editing (DaVinci Resolve); Productivity: Average tokens/second for LLM inference (Llama 3.1 8B, ISL/OSL = 2000/100, quantized to int4); Development: Average req/second for transformer fine tuning using PyTorch (BERT, BS=32, mixed precision). **AI-Enhanced Apps** ## All Your Favorite Apps, Now Smarter More and more apps are featuring AI every day—and with an RTX AI PC, you’ll be able to power them in the best possible way. LOGO CRAWL v6 [Explore Apps](https://www.nvidia.com/en-us/gpu-accelerated-applications/?filter=eyJOVklESUFBSUFjY2VsZXJhdGVkUHJvZ3JhbSI6WyJBY2NlcHRlZCAtIEFwcHJvdmVkIGxvZ28gdXNlIl19) AI That TOPS the Charts Measured in trillions of operations per second, or TOPS, the AI processing power of GeForce RTX Laptop GPUs is in a class all its own. 1. GeForce RTX 50 Series 2. GeForce RTX 40 Series | | | **GeForce RTX 5090 Laptop GPU** | **GeForce RTX 5080 Laptop GPU** | **GeForce RTX 5070 Ti Laptop GPU** | **GeForce RTX 5070 Laptop GPU** | **GeForce RTX 5060 Laptop GPU** | | --- | --- | --- | --- | --- | --- | --- | | | **AI Performance** | 1824 AI TOPS | 1334 AI TOPS | 992 AI TOPS | 798 AI TOPS | 572 AI TOPS | | | | **GeForce RTX 4090 Laptop GPU** | **GeForce RTX 4080 Laptop GPU** | **GeForce RTX 4070 Ti Laptop GPU** | **GeForce RTX 4070 Laptop GPU** | **GeForce RTX 4050 Laptop GPU** | | --- | --- | --- | --- | --- | --- | --- | | | **AI Performance** | 686 AI TOPS | 542 AI TOPS | 321 AI TOPS | 233 AI TOPS | 194 AI TOPS | ## Shop Now Find the perfect RTX AI PC desktops and laptops to level up everything you do. ## Shop All GeForce RTX ### GeForce RTX 50 Series Graphics Cards [See All Buying Options](https://marketplace.nvidia.com/en-gb/consumer/graphics-cards/?locale=en-gb&page=1&limit=12&gpu=RTX%205090,RTX%205080,RTX%205070%20Ti,RTX%205070,RTX%205060%20Ti) [Learn More](https://www.nvidia.com/en-gb/geforce/graphics-cards/50-series.md) ### GeForce RTX 50 Series Desktops [See All Buying Options](https://marketplace.nvidia.com/en-gb/consumer/gaming-desktops/?locale=en-gb&page=1&limit=12&category=DESKTOP&gpu=RTX%205090,RTX%205080,RTX%205070%20Ti,RTX%205070,RTX%205060%20Ti) [Learn More](https://www.nvidia.com/en-gb/geforce/graphics-cards/50-series.md) ### GeForce RTX 50 Series Laptops [See All Buying Options](https://marketplace.nvidia.com/en-gb/consumer/gaming-laptops/?locale=en-gb&page=1&limit=12&gpu=RTX%205090,RTX%205080,RTX%205070%20Ti) [Learn More](https://www.nvidia.com/en-gb/geforce/laptops/50-series.md) ### NVIDIA Studio Laptops and Desktops [See All Buying Options](https://marketplace.nvidia.com/en-gb/consumer/studio-laptops/?locale=en-us&page=1&limit=12&gpu=RTX%205090,RTX%205080,RTX%205070%20Ti,RTX%205070,RTX%205060,RTX%205050) [Learn More](https://www.nvidia.com/en-us/studio/help-me-choose.md) ### GeForce RTX 30 Series Graphics Cards and Desktops [See All Buying Options](https://store.nvidia.com/en-gb/geforce/store/?page=1&limit=9&locale=en-gb&category=GPU,DESKTOP&gpu=RTX%203090%20Ti,RTX%203090,RTX%203080%20Ti,RTX%203080,RTX%203070,RTX%203070%20Ti,RTX%203060%20Ti,RTX%203060,RTX%203050,RTX%203050%20Ti) [Learn More](https://www.nvidia.com/en-gb/geforce/graphics-cards/30-series.md) ### GeForce RTX 30 Series Laptops [See All Buying Options](https://store.nvidia.com/en-gb/geforce/store/laptop/?page=1&limit=9&locale=en-gb&category=LAPTOP&gpu=RTX%203090%20Ti,RTX%203090,RTX%203080%20Ti,RTX%203080,RTX%203070,RTX%203070%20Ti,RTX%203060%20Ti,RTX%203060,RTX%203050,RTX%203050%20Ti) [Learn More](https://www.nvidia.com/en-gb/geforce/laptops/30-series.md) Sign up for the latest NVIDIA RTX AI news. [Subscribe](#subscribe-ai) ![YouTube Video](https://img.youtube.com/vi_webp/K_MnB6rrlhA/maxresdefault.webp) Consent for Optional Cookies (googleCookiePolicyLink)YouTube sets performance, advertising, and other optional cookies(/googleCookiePolicyLink) when you watch embedded videos. To watch this video, you need to turn on optional cookies for the site. By clicking “Accept and Play Video,” you will automatically turn on advertising and other optional cookies for the site and accept our (nvidiaTermsOfServiceLink)Terms of Service(/nvidiaTermsOfServiceLink) (which contains important waivers). Please see our (nvidiaPrivacyPolicyLink)Privacy Policy(/nvidiaPrivacyPolicyLink) and (nvidiaCookiePolicyLink)Cookie Policy(/nvidiaCookiePolicyLink) for more information. Cancel Accept and Play Video Alternatively, you can (youtubeLink)watch this video on YouTube(/youtubeLink). ![YouTube Video](https://img.youtube.com/vi_webp/nR-vP_7XFHE/maxresdefault.webp) Consent for Optional Cookies (googleCookiePolicyLink)YouTube sets performance, advertising, and other optional cookies(/googleCookiePolicyLink) when you watch embedded videos. To watch this video, you need to turn on optional cookies for the site. By clicking “Accept and Play Video,” you will automatically turn on advertising and other optional cookies for the site and accept our (nvidiaTermsOfServiceLink)Terms of Service(/nvidiaTermsOfServiceLink) (which contains important waivers). Please see our (nvidiaPrivacyPolicyLink)Privacy Policy(/nvidiaPrivacyPolicyLink) and (nvidiaCookiePolicyLink)Cookie Policy(/nvidiaCookiePolicyLink) for more information. Cancel Accept and Play Video Alternatively, you can (youtubeLink)watch this video on YouTube(/youtubeLink). ![YouTube Video](https://img.youtube.com/vi_webp/vrkXR2ZjGGI/maxresdefault.webp) Consent for Optional Cookies (googleCookiePolicyLink)YouTube sets performance, advertising, and other optional cookies(/googleCookiePolicyLink) when you watch embedded videos. To watch this video, you need to turn on optional cookies for the site. By clicking “Accept and Play Video,” you will automatically turn on advertising and other optional cookies for the site and accept our (nvidiaTermsOfServiceLink)Terms of Service(/nvidiaTermsOfServiceLink) (which contains important waivers). Please see our (nvidiaPrivacyPolicyLink)Privacy Policy(/nvidiaPrivacyPolicyLink) and (nvidiaCookiePolicyLink)Cookie Policy(/nvidiaCookiePolicyLink) for more information. Cancel Accept and Play Video Alternatively, you can (youtubeLink)watch this video on YouTube(/youtubeLink). ![YouTube Video](https://img.youtube.com/vi_webp/DgATJ38k8-M/maxresdefault.webp) Consent for Optional Cookies (googleCookiePolicyLink)YouTube sets performance, advertising, and other optional cookies(/googleCookiePolicyLink) when you watch embedded videos. To watch this video, you need to turn on optional cookies for the site. By clicking “Accept and Play Video,” you will automatically turn on advertising and other optional cookies for the site and accept our (nvidiaTermsOfServiceLink)Terms of Service(/nvidiaTermsOfServiceLink) (which contains important waivers). Please see our (nvidiaPrivacyPolicyLink)Privacy Policy(/nvidiaPrivacyPolicyLink) and (nvidiaCookiePolicyLink)Cookie Policy(/nvidiaCookiePolicyLink) for more information. Cancel Accept and Play Video Alternatively, you can (youtubeLink)watch this video on YouTube(/youtubeLink). ![YouTube Video](https://img.youtube.com/vi_webp/cSSSn10HgZA/maxresdefault.webp) Consent for Optional Cookies (googleCookiePolicyLink)YouTube sets performance, advertising, and other optional cookies(/googleCookiePolicyLink) when you watch embedded videos. To watch this video, you need to turn on optional cookies for the site. By clicking “Accept and Play Video,” you will automatically turn on advertising and other optional cookies for the site and accept our (nvidiaTermsOfServiceLink)Terms of Service(/nvidiaTermsOfServiceLink) (which contains important waivers). Please see our (nvidiaPrivacyPolicyLink)Privacy Policy(/nvidiaPrivacyPolicyLink) and (nvidiaCookiePolicyLink)Cookie Policy(/nvidiaCookiePolicyLink) for more information. Cancel Accept and Play Video Alternatively, you can (youtubeLink)watch this video on YouTube(/youtubeLink). ## Get the Latest AI on RTX News Welcome back. Not you? Log Out Welcome back. Not you? Clear form