The hardest part about finding games like Hollow Knight is knowing where to start. The overwhelming success of Team Cherry's award-winning 2017 game--and anticipation of its long-awaited sequel, Silksong--prompted a flood of similar games all looking to capture the magic of combining soulslike combat with deep exploration. Some are more inventive than others, building on Hollow Knight's foundations to push that style of game forward in new or unexpected ways. Others take a specific aspect, such as grueling boss fights, and run with it. We've combed through the lot and picked out 15 of the best games like Hollow Knight to get you started in this impressively varied sub-genre.If you're not as excited about combat and want puzzles and exploration instead, head over to our list of the best metroidvania games. Nine Sols Platforms: PC, Xbox Series X|S, PlayStation 5Release Date: May 29, 2024Developer: Red Candle GamesIf Hollow Knight is the Dark Souls of metroidvanias, then Nine Sols is the genre's Sekiro: Shadows Die Twice. Parrying is at the core of almost everything you do in Nine Sols, from dealing with standard enemies to wearing down some of its relentless bosses. Among games like Hollow Knight, it's also one of the most thematically and visually distinct. Developer Red Candle Games call Nine Sols a "Taopunk," a blend of sci-fi punk with traditional Taoist architecture and symbolism. Most protagonists in games like these are blank slates, but Nine Sols adds a personal touch by making the personality of its hero, Yi, an important part of the story. Yi starts out seeking revenge, and ends up on a journey to save the world and himself, becoming a reluctant hero in the process. See at Humble Animal Well Platforms: PC, Nintendo Switch, Xbox Series X|S, PlayStation 5Release Date: May 9, 2024Developer: Billy Basso, Shared Memory LLCAnimal Well is a puzzle, or more accurately, a lot of puzzles. There's a bit of combat and some platforming, but mostly, it's about trying to unravel dozens of mysteries big and small as you delve ever further into a maze that wouldn't be out of place in Lewis Carrol's Wonderland stories. Explaining too much about what's going on would spoil what makes Animal Well special, but the most interesting and even subversive parts of it is that you have almost nothing to guide you and can make discoveries in any order. That freedom creates a sense of discovery and wonder that's often absent from the procedural methods inherent in these kinds of games.Read our Animal Well review. See at Steam Ultros Platforms: PC, PlayStation 5Release Date: February 13, 2024Developer: HadoqueOf all the games like Hollow Knight, Ultros takes the most organic approach to metroidvanias, and we mean that literally. You, an intergalactic explorer, arrive on a psychedelic space colony called The Sarcophagus and find it teeming with exotic life and mysterious spiritual energies. You use the life force and remains of enemies to nourish your mind and unlock new abilities, and there's a scoring system that ranks how efficiently you defeat your foes. That determines the quality of the loot they drop, so if you want to unlock and improve your skills, you have to plan each encounter carefully. Ultros is also absolutely beautiful, a dream-like blend of esoteric architecture and wild ecosystems with closer ties to the Sarcophagus' secrets than Ultros initially suggests. Read our Ultros review. See at Fanatical Blasphemous 2 Platforms: PlayStation 5, PC, Xbox Series X|S, Nintendo SwitchRelease Date: August 24, 2023Developer: The Game KitchenBlasphemous 2's big addition over its predecessor--apart from even more ghoulish and gory moments--is the inclusion of more platforming. The first Blasphemous is a bit one-note, which is great if you're just here for the combat, but not so much if you want, well, anything else. Blasphemous 2 throws in some challenging and smartly-designed platforming as well, bringing it closer to the likes of Hollow Knight. Better still, developer The Game Kitchen was more ambitious with its environment design as well, with more complex layouts, better backgrounds and lighting, and even colors that aren't brown, grey, and blood.Read our Blasphemous 2 review. See at Fanatical Elden Ring Platforms: PlayStation 5, PlayStation 4, Xbox One, Xbox Series X|S, PCRelease Date: February 25, 2022Developer: FromSoftwareOkay, so Elden Ring doesn't have the exploration style of Hollow Knight, but it does have the kind of grueling combat that inspired Team Cherry's spectacular boss fights, and lots of it. It's FromSoftware's first open-world game, one that follows a lone, nameless warrior in their bid to bring salvation to a shattered land--or make its ruin everlasting. Mostly, though, it's a giant playground for dozens of exceptionally well-designed and challenging bosses to stomp (you) around in, with enemies ranging from fire-spewing land dragons to the spirits of an ancient civilization and a gigantic, greatsword-wielding prince on his favorite little horsey. Read our Elden Ring review. See at Fanatical Cuphead Platforms: PlayStation 4, Xbox One, PCRelease Date: September 29, 2017Developer: Studio MDHRIf you really like boss fights and are less bothered about exploration and all the other Hollow Knight-y bits, Cuphead is definitely worth checking out. Don't let Studio MDHR's retro cartoon style give you the wrong impression, either. Cuphead's cutesy bosses demand careful planning, precise execution, and a lot of patience. The battles aren't the only thing Cuphead has going for it, though. Studio MDHR's exquisite animation, the soundtrack, even period-specific cartoon-style sound effects--the entire game is a spectacle in the best way. Read our Cuphead review. See at Fanatical Castlevania Advance Collection Platforms: Xbox Series X|S, Nintendo Switch, PlayStation 5, PCRelease Date: September 23, 2021Developer: KonamiAny of the Castlevania bundles are strong picks, but the Castlevania Advance Collection should be your go-to choice for the kind of classic action that Hollow Knight builds on. The collection includes Circle of the Moon--low on our list of the best Castlevania games only on account of it not really doing anything that Symphony of the Night didn't--the excellent Harmony of Dissonance, and the even better Aria of Sorrow. Sorrow is the standout inclusion, one that radically shook up the Castlevania formula by removing the Belmonts from the equation, telling an entirely new story set in the distant (at the time) future, and giving the protagonist, Soma, an ability that absorbed enemy souls for use in combat. It helps that Sorrow, as well as Harmony, have an excellent selection of bosses and some fantastically moody settings, too. See at Humble Ori and the Blind Forest Platforms: PlayStation 4, Xbox One, Nintendo Switch, PCRelease Date: March 11, 2015Developer: Moon StudiosOri and the Blind Forest starts like the end of a Disney movie. A cute little creature finds a family in the middle of a dream-like forest, and that family gets taken away from them. Your job is to figure out why and find a way to save the woods. Ori takes Hollow Knight's demanding platforming even further with some segments that wouldn't feel out of place in something like Celeste, but the real stand-out feature is the map. In addition to being a well-designed metroidvania world, it's absolutely gorgeous and a delight to explore. Blind Forest is a modern classic, and its sequel, Ori and the Will of the Wisps manages to improve it even further.Read our Ori and the Blind Forest review and Ori and the Will of the Wisps review. See at Fanatical Ender Magnolia: Bloom in the Mist Platforms: PlayStation 4, PlayStation 5, Xbox One, Xbox Series X|S, PC, Nintendo SwitchRelease Date: March 25, 2024Developer: AdiglobeIn the onslaught of games like Hollow Knight that released following Team Cherry's success, developer Adiglobe decided to shake up the formula first with Ender Lilies and then with the more refined Ender Magnolia. You've got your standard elements, such as gigantic bosses that force you to learn their patterns and a puzzle-like map that unfolds as you gain more powers. Those powers, however, are the spirits of fallen friends who also aid you in combat. You find several, but can only recruit a handful at a time, which adds a layer of strategy to exploration and combat. There's also a strong sense of emotional attachment, since you and your ghostly allies have history and connection of a kind that's often missing in these games when you just play as an outside observer. See at Steam Metroid Dread Platforms: Nintendo SwitchRelease Date: October 8, 2021Developer: Nintendo EADAny Metroid game is going to have something of that Hollow Knight feel, since the sci-fi series is a big part of where the genre and Hollow Knight in particular came from. However, the easiest to get your hands on without having to pay for a subscription is Metroid Dread. It's the culmination of the 2D Metroid saga that started in 1986, but you can get by just fine if this is your first. Dread follows bounty hunter Samus Aran as she searches for evidence of a deadly parasite, only to find herself hunted by rogue robots--sometimes. The stealth segments that have you hiding from your metal hunters only take place in specific areas. The rest of Metroid Dread takes you across sprawling subterranean research labs, sunken testing stations, and extravagant dwellings, featuring the series' biggest map ever, stuffed with secrets and formidable bosses. If you really enjoy those bosses, Dread has a boss rush mode you can test yourself against as well.Read our Metroid Dread review. See at Amazon Lone Fungus Platforms: PCRelease Date: September 21, 2021Developer: Basti GamesThese types of games tend to lean more toward the edgy, dark, and broody side of things, which makes Lone Fungus a gem in the genre. You're the last mushroom on Earth, exploring a vast network of tunnels and temples in search of treasure and using magic skills that change form depending on how you swing your sword. A ball of energy is damaging to one enemy, for example, but you can smack it and shatter it into several projectiles to clear out lots of foes at once. Best of all, though, Lone Fungus has a robust Assist Mode that lets people of all skill levels enjoy the game and just makes it more relaxed in general, with features such as extra platforms, no costs for spells, slower platforms, and invincibility so your little fungus won't die. See at Humble Dead Cells Platforms: PlayStation 4, Xbox One, Nintendo Switch, PCRelease Date: May 10, 2027Developer: Motion TwinDead Cells throws roguelite randomness into the metroidvania mix and ratchets up the challenge as well. You're a spirit determined to figure out why you died, and in the absence of a tangible vessel for your ethereal self, you pilot shambling corpses in a bid to make it through streets, swamps, dungeons, and horrors untold. These corpses aren't the sturdiest, so when you fail, they fall apart and you start again. Eventually, you can unlock permanent upgrades, but with no checkpoints at any stage of the journey, you'll have to rely on your skill with weapons and knowledge of enemy behavior to make it through. If you enjoy Dead Cells, you can pick up DLC packs that add new locations, weapons, and enemies, and there's even a Castlevania-themed expansion as well. Read our Dead Cells review. See at Fanatical Monster Sanctuary Platforms: Nintendo Switch, Xbox Series One, PCRelease Date: August 28, 2019Developer: DeveloperIf you like a bit of Pokemon with your Hollow Knight, Monster Sanctuary might be for you. You play as a fledgling monster tamer out to explore the vast world with just a single critter by your side. You'll find and tame more, training them into the best versions of themselves and using their abilities not just to deal with threats in the sanctuary, but to explore its secrets and hidden areas as well. Monster Sanctuary is lighter and breezier than some games on this list, but if something more intense is to your liking, there's a robust PvP element where you can challenge other players and their monster teams too. See at Fanatical Salt and Sanctuary Platforms: PC, PlayStation 4, Xbox One, Nintendo SwitchRelease Date: March 15, 2016Developer: Ska StudiosUpgrades and customization are usually rather limited in games like Hollow Knight, which makes Salt and Sanctuary, with its more in-depth RPG components, something special. You play as a sailor, washed up on some evil-looking, godforsaken island and foolish enough to go exploring the mysterious labyrinth underneath. What you find is a parade of nightmares and some spectacular, bone-crunchingly hard 2D boss fights that are among the best Soulslike challenges out there. Read our Salt and Sanctuary review. See at Humble Bō: Path of the Teal Lotus Platforms: PC, Xbox Series X|S, PlayStation 5, Nintendo SwitchRelease Date: July 17, 2024Developer: Squid Shock StudiosMovement is often a means to an end in video games--a double jump that propels you higher, for example, or a dash that lets you avoid dangerous terrain. In Bō: Path of the Teal Lotus, it's no exaggeration to say movement is everything. Bouncing, flying, gliding, moving with magical speed, and navigating the hand-drawn world is just as important as battling the myths and monsters inspired by Japanese folklore. Path of the Teal Lotus is one of the most elegant platformers around, and it even has a reset system where you can pause or rewind a failed jump to try again, perfect for learning some of the more difficult segments. Read our Path of the Teal Lotus review. See at Fanatical

As online hangouts expand, Discord wants to be developers' easy on-ramp for community-building and socialization.

Urban populations are expected to double by 2050, which means around 2.5 billion people could be added to urban areas by the middle of the century, driving the need for more sustainable urban planning and public services. Cities across the globe are turning to digital twins and AI agents for urban planning scenario analysis and data-driven operational decisions. Building a digital twin of a city and testing smart city AI agents within it, however, is a complex and resource-intensive endeavor, fraught with technical and operational challenges. To address those challenges, NVIDIA today announced the NVIDIA Omniverse Blueprint for smart city AI, a reference framework that combines the NVIDIA Omniverse, Cosmos, NeMo and Metropolis platforms to bring the benefits of physical AI to entire cities and their critical infrastructure. Using the blueprint, developers can build simulation-ready, or SimReady, photorealistic digital twins of cities to build and test AI agents that can help monitor and optimize city operations. Leading companies including XXII, AVES Reality, Akila, Blyncsy, Bentley, Cesium, K2K, Linker Vision, Milestone Systems, Nebius, SNCF Gares&Connexions, Trimble and Younite AI are among the first to use the new blueprint. NVIDIA Omniverse Blueprint for Smart City AI  The NVIDIA Omniverse Blueprint for smart city AI provides the complete software stack needed to accelerate the development and testing of AI agents in physically accurate digital twins of cities. It includes: NVIDIA Omniverse to build physically accurate digital twins and run simulations at city scale. NVIDIA Cosmos to generate synthetic data at scale for post-training AI models. NVIDIA NeMo to curate high-quality data and use that data to train and fine-tune vision language models (VLMs) and large language models. NVIDIA Metropolis to build and deploy video analytics AI agents based on the NVIDIA AI Blueprint for video search and summarization (VSS), helping process vast amounts of video data and provide critical insights to optimize business processes. The blueprint workflow comprises three key steps. First, developers create a SimReady digital twin of locations and facilities using aerial, satellite or map data with Omniverse and Cosmos. Second, they can train and fine-tune AI models, like computer vision models and VLMs, using NVIDIA TAO and NeMo Curator to improve accuracy for vision AI use cases​. Finally, real-time AI agents powered by these customized models are deployed to alert, summarize and query camera and sensor data using the Metropolis VSS blueprint. NVIDIA Partner Ecosystem Powers Smart Cities Worldwide The blueprint for smart city AI enables a large ecosystem of partners to use a single workflow to build and activate digital twins for smart city use cases, tapping into a combination of NVIDIA’s technologies and their own. SNCF Gares&Connexions, which operates a network of 3,000 train stations across France and Monaco, has deployed a digital twin and AI agents to enable real-time operational monitoring, emergency response simulations and infrastructure upgrade planning. This helps each station analyze operational data such as energy and water use, and enables predictive maintenance capabilities, automated reporting and GDPR-compliant video analytics for incident detection and crowd management. Powered by Omniverse, Metropolis and solutions from ecosystem partners Akila and XXII, SNCF Gares&Connexions’ physical AI deployment at the Monaco-Monte-Carlo and Marseille stations has helped SNCF Gares&Connexions achieve a 100% on-time preventive maintenance completion rate, a 50% reduction in downtime and issue response time, and a 20% reduction in energy consumption. https://blogs.nvidia.com/wp-content/uploads/2025/06/01-Monaco-Akila.mp4 The city of Palermo in Sicily is using AI agents and digital twins from its partner K2K to improve public health and safety by helping city operators process and analyze footage from over 1,000 public video streams at a rate of nearly 50 billion pixels per second. Tapped by Sicily, K2K’s AI agents — built with the NVIDIA AI Blueprint for VSS and cloud solutions from Nebius — can interpret and act on video data to provide real-time alerts on public events. To accurately predict and resolve traffic incidents, K2K is generating synthetic data with Cosmos world foundation models to simulate different driving conditions. Then, K2K uses the data to fine-tune the VLMs powering the AI agents with NeMo Curator. These simulations enable K2K’s AI agents to create over 100,000 predictions per second. https://blogs.nvidia.com/wp-content/uploads/2025/06/02-K2K-Polermo-1600x900-1.mp4 Milestone Systems — in collaboration with NVIDIA and European cities — has launched Project Hafnia, an initiative to build an anonymized, ethically sourced video data platform for cities to develop and train AI models and applications while maintaining regulatory compliance. Using a combination of Cosmos and NeMo Curator on NVIDIA DGX Cloud and Nebius’ sovereign European cloud infrastructure, Project Hafnia scales up and enables European-compliant training and fine-tuning of video-centric AI models, including VLMs, for a variety of smart city use cases. The project’s initial rollout, taking place in Genoa, Italy, features one of the world’s first VLM models for intelligent transportation systems. https://blogs.nvidia.com/wp-content/uploads/2025/06/03-Milestone.mp4 Linker Vision was among the first to partner with NVIDIA to deploy smart city digital twins and AI agents for Kaohsiung City, Taiwan — powered by Omniverse, Cosmos and Metropolis. Linker Vision worked with AVES Reality, a digital twin company, to bring aerial imagery of cities and infrastructure into 3D geometry and ultimately into SimReady Omniverse digital twins. Linker Vision’s AI-powered application then built, trained and tested visual AI agents in a digital twin before deployment in the physical city. Now, it’s scaling to analyze 50,000 video streams in real time with generative AI to understand and narrate complex urban events like floods and traffic accidents. Linker Vision delivers timely insights to a dozen city departments through a single integrated AI-powered platform, breaking silos and reducing incident response times by up to 80%. https://blogs.nvidia.com/wp-content/uploads/2025/06/02-Linker-Vision-1280x680-1.mp4 Bentley Systems is joining the effort to bring physical AI to cities with the NVIDIA blueprint. Cesium, the open 3D geospatial platform, provides the foundation for visualizing, analyzing and managing infrastructure projects and ports digital twins to Omniverse. The company’s AI platform Blyncsy uses synthetic data generation and Metropolis to analyze road conditions and improve maintenance. Trimble, a global technology company that enables essential industries including construction, geospatial and transportation, is exploring ways to integrate components of the Omniverse blueprint into its reality capture workflows and Trimble Connect digital twin platform for surveying and mapping applications for smart cities. Younite AI, a developer of AI and 3D digital twin solutions, is adopting the blueprint to accelerate its development pipeline, enabling the company to quickly move from operational digital twins to large-scale urban simulations, improve synthetic data generation, integrate real-time IoT sensor data and deploy AI agents. Learn more about the NVIDIA Omniverse Blueprint for smart city AI by attending this GTC Paris session or watching the on-demand video after the event. Sign up to be notified when the blueprint is available. Watch the NVIDIA GTC Paris keynote from NVIDIA founder and CEO Jensen Huang at VivaTech, and explore GTC Paris sessions.

Telecom companies last year spent nearly $295 billion in capital expenditures and over $1 trillion in operating expenditures. These large expenses are due in part to laborious manual processes that telcos face when operating networks that require continuous optimizations. For example, telcos must constantly tune network parameters for tasks — such as transferring calls from one network to another or distributing network traffic across multiple servers — based on the time of day, user behavior, mobility and traffic type. These factors directly affect network performance, user experience and energy consumption. To automate these optimization processes and save costs for telcos across the globe, NVIDIA today unveiled at GTC Paris its first AI Blueprint for telco network configuration. At the blueprint’s core are customized large language models trained specifically on telco network data — as well as the full technical and operational architecture for turning the LLMs into an autonomous, goal-driven AI agent for telcos. Automate Network Configuration With the AI Blueprint NVIDIA AI Blueprints — available on build.nvidia.com — are customizable AI workflow examples. They include reference code, documentation and deployment tools that show enterprise developers how to deliver business value with NVIDIA NIM microservices. The AI Blueprint for telco network configuration — built with BubbleRAN 5G solutions and datasets — enables developers, network engineers and telecom providers to automatically optimize the configuration of network parameters using agentic AI. This can streamline operations, reduce costs and significantly improve service quality by embedding continuous learning and adaptability directly into network infrastructures. Traditionally, network configurations required manual intervention or followed rigid rules to adapt to dynamic network conditions. These approaches limited adaptability and increased operational complexities, costs and inefficiencies. The new blueprint helps shift telco operations from relying on static, rules-based systems to operations based on dynamic, AI-driven automation. It enables developers to build advanced, telco-specific AI agents that make real-time, intelligent decisions and autonomously balance trade-offs — such as network speed versus interference, or energy savings versus utilization — without human input. Powered and Deployed by Industry Leaders Trained on 5G data generated by BubbleRAN, and deployed on the BubbleRAN 5G O-RAN platform, the blueprint provides telcos with insight on how to set various parameters to reach performance goals, like achieving a certain bitrate while choosing an acceptable signal-to-noise ratio — a measure that impacts voice quality and thus user experience. With the new AI Blueprint, network engineers can confidently set initial parameter values and update them as demanded by continuous network changes. Norway-based Telenor Group, which serves over 200 million customers globally, is the first telco to integrate the AI Blueprint for telco network configuration as part of its initiative to deploy intelligent, autonomous networks that meet the performance and agility demands of 5G and beyond. “The blueprint is helping us address configuration challenges and enhance quality of service during network installation,” said Knut Fjellheim, chief technology innovation officer at Telenor Maritime. “Implementing it is part of our push toward network automation and follows the successful deployment of agentic AI for real-time network slicing in a private 5G maritime use case.” Industry Partners Deploy Other NVIDIA-Powered Autonomous Network Technologies The AI Blueprint for telco network configuration is just one of many announcements at NVIDIA GTC Paris showcasing how the telecom industry is using agentic AI to make autonomous networks a reality. Beyond the blueprint, leading telecom companies and solutions providers are tapping into NVIDIA accelerated computing, software and microservices to provide breakthrough innovations poised to vastly improve networks and communications services — accelerating the progress to autonomous networks and improving customer experiences. NTT DATA is powering its agentic platform for telcos with NVIDIA accelerated compute and the NVIDIA AI Enterprise software platform. Its first agentic use case is focused on network alarms management, where NVIDIA NIM microservices help automate and power observability, troubleshooting, anomaly detection and resolution with closed loop ticketing. Tata Consultancy Services is delivering agentic AI solutions for telcos built on NVIDIA DGX Cloud and using NVIDIA AI Enterprise to develop, fine-tune and integrate large telco models into AI agent workflows. These range from billing and revenue assurance, autonomous network management to hybrid edge-cloud distributed inference. For example, the company’s anomaly management agentic AI model includes real-time detection and resolution of network anomalies and service performance optimization. This increases business agility and improves operational efficiencies by up to 40% by eliminating human intensive toils, overheads and cross-departmental silos. Prodapt has introduced an autonomous operations workflow for networks, powered by NVIDIA AI Enterprise, that offers agentic AI capabilities to support autonomous telecom networks. AI agents can autonomously monitor networks, detect anomalies in real time, initiate diagnostics, analyze root causes of issues using historical data and correlation techniques, automatically execute corrective actions, and generate, enrich and assign incident tickets through integrated ticketing systems. Accenture announced its new portfolio of agentic AI solutions for telecommunications through its AI Refinery platform, built on NVIDIA AI Enterprise software and accelerated computing. The first available solution, the NOC Agentic App, boosts network operations center tasks by using a generative AI-driven, nonlinear agentic framework to automate processes such as incident and fault management, root cause analysis and configuration planning. Using the Llama 3.1 70B NVIDIA NIM microservice and the AI Refinery Distiller Framework, the NOC Agentic App orchestrates networks of intelligent agents for faster, more efficient decision-making. Infosys is announcing its agentic autonomous operations platform, called Infosys Smart Network Assurance (ISNA), designed to accelerate telecom operators’ journeys toward fully autonomous network operations. ISNA helps address long-standing operational challenges for telcos — such as limited automation and high average time to repair — with an integrated, AI-driven platform that reduces operational costs by up to 40% and shortens fault resolution times by up to 30%. NVIDIA NIM and NeMo microservices enhance the platform’s reasoning and hallucination-detection capabilities, reduce latency and increase accuracy. Get started with the new blueprint today. Learn more about the latest AI advancements for telecom and other industries at NVIDIA GTC Paris, running through Thursday, June 12, at VivaTech, including a keynote from NVIDIA founder and CEO Jensen Huang and a special address from Ronnie Vasishta, senior vice president of telecom at NVIDIA. Plus, hear from industry leaders in a panel session with Orange, Swisscom, Telenor and NVIDIA.

Project G-Assist — available through the NVIDIA App — is an experimental AI assistant that helps tune, control and optimize NVIDIA GeForce RTX systems. NVIDIA’s Plug and Play: Project G-Assist Plug-In Hackathon — running virtually through Wednesday, July 16 — invites the community to explore AI and build custom G-Assist plug-ins for a chance to win prizes and be featured on NVIDIA social media channels. G-Assist allows users to control their RTX GPU and other system settings using natural language, thanks to a small language model that runs on device. It can be used from the NVIDIA Overlay in the NVIDIA App without needing to tab out or switch programs. Users can expand its capabilities via plug-ins and even connect it to agentic frameworks such as Langflow. Below, find popular G-Assist plug-ins, hackathon details and tips to get started. Plug-In and Win Join the hackathon by registering and checking out the curated technical resources. G-Assist plug-ins can be built in several ways, including with Python for rapid development, with C++ for performance-critical apps and with custom system interactions for hardware and operating system automation. For those that prefer vibe coding, the G-Assist Plug-In Builder — a ChatGPT-based app that allows no-code or low-code development with natural language commands — makes it easy for enthusiasts to start creating plug-ins. To submit an entry, participants must provide a GitHub repository, including source code file (plugin.py), requirements.txt, manifest.json, config.json (if applicable), a plug-in executable file and READme code. Then, submit a video — between 30 seconds and two minutes — showcasing the plug-in in action. Finally, hackathoners must promote their plug-in using #AIonRTXHackathon on a social media channel: Instagram, TikTok or X. Submit projects via this form by Wednesday, July 16. Judges will assess plug-ins based on three main criteria: 1) innovation and creativity, 2) technical execution and integration, reviewing technical depth, G-Assist integration and scalability, and 3) usability and community impact, aka how easy it is to use the plug-in. Winners will be selected on Wednesday, Aug. 20. First place will receive a GeForce RTX 5090 laptop, second place a GeForce RTX 5080 GPU and third a GeForce RTX 5070 GPU. These top three will also be featured on NVIDIA’s social media channels, get the opportunity to meet the NVIDIA G-Assist team and earn an NVIDIA Deep Learning Institute self-paced course credit. Project G-Assist requires a GeForce RTX 50, 40 or 30 Series Desktop GPU with at least 12GB of VRAM, Windows 11 or 10 operating system, a compatible CPU (Intel Pentium G Series, Core i3, i5, i7 or higher; AMD FX, Ryzen 3, 5, 7, 9, Threadripper or higher), specific disk space requirements and a recent GeForce Game Ready Driver or NVIDIA Studio Driver. Plug-In(spiration) Explore open-source plug-in samples available on GitHub, which showcase the diverse ways on-device AI can enhance PC and gaming workflows. Popular plug-ins include: Google Gemini: Enables search-based queries using Google Search integration and large language model-based queries using Gemini capabilities in real time without needing to switch programs from the convenience of the NVIDIA App Overlay. Discord: Enables users to easily share game highlights or messages directly to Discord servers without disrupting gameplay. IFTTT: Lets users create automations across hundreds of compatible endpoints to trigger IoT routines — such as adjusting room lights and smart shades, or pushing the latest gaming news to a mobile device. Spotify: Lets users control Spotify using simple voice commands or the G-Assist interface to play favorite tracks and manage playlists. Twitch: Checks if any Twitch streamer is currently live and can access detailed stream information such as titles, games, view counts and more. Get G-Assist(ance)  Join the NVIDIA Developer Discord channel to collaborate, share creations and gain support from fellow AI enthusiasts and NVIDIA staff. Save the date for NVIDIA’s How to Build a G-Assist Plug-In webinar on Wednesday, July 9, from 10-11 a.m. PT, to learn more about Project G-Assist capabilities, discover the fundamentals of building, testing and deploying Project G-Assist plug-ins, and participate in a live Q&A session. Explore NVIDIA’s GitHub repository, which provides everything needed to get started developing with G-Assist, including sample plug-ins, step-by-step instructions and documentation for building custom functionalities. Learn more about the ChatGPT Plug-In Builder to transform ideas into functional G-Assist plug-ins with minimal coding. The tool uses OpenAI’s custom GPT builder to generate plug-in code and streamline the development process. NVIDIA’s technical blog walks through the architecture of a G-Assist plug-in, using a Twitch integration as an example. Discover how plug-ins work, how they communicate with G-Assist and how to build them from scratch. Each week, the RTX AI Garage blog series features community-driven AI innovations and content for those looking to learn more about NVIDIA NIM microservices and AI Blueprints, as well as building AI agents, creative workflows, digital humans, productivity apps and more on AI PCs and workstations.  Plug in to NVIDIA AI PC on Facebook, Instagram, TikTok and X — and stay informed by subscribing to the RTX AI PC newsletter. Follow NVIDIA Workstation on LinkedIn and X.  See notice regarding software product information.

From the heart of Germany’s automotive sector to manufacturing hubs across France and Italy, Europe is embracing industrial AI and advanced AI-powered robotics to address labor shortages, boost productivity and fuel sustainable economic growth. Robotics companies are developing humanoid robots and collaborative systems that integrate AI into real-world manufacturing applications. Supported by a $200 billion investment initiative and coordinated efforts from the European Commission, Europe is positioning itself at the forefront of the next wave of industrial automation, powered by AI. This momentum is on full display at Automatica — Europe’s premier conference on advancements in robotics, machine vision and intelligent manufacturing — taking place this week in Munich, Germany. NVIDIA and its ecosystem of partners and customers are showcasing next-generation robots, automation and AI technologies designed to accelerate the continent’s leadership in smart manufacturing and logistics. NVIDIA Technologies Boost Robotics Development  Central to advancing robotics development is Europe’s first industrial AI cloud, announced at NVIDIA GTC Paris at VivaTech earlier this month. The Germany-based AI factory, featuring 10,000 NVIDIA GPUs, provides European manufacturers with secure, sovereign and centralized AI infrastructure for industrial workloads. It will support applications ranging from design and engineering to factory digital twins and robotics. To help accelerate humanoid development, NVIDIA released NVIDIA Isaac GR00T N1.5 — an open foundation model for humanoid robot reasoning and skills. This update enhances the model’s adaptability and ability to follow instructions, significantly improving its performance in material handling and manufacturing tasks. To help post-train GR00T N1.5, NVIDIA has also released the Isaac GR00T-Dreams blueprint — a reference workflow for generating vast amounts of synthetic trajectory data from a small number of human demonstrations — enabling robots to generalize across behaviors and adapt to new environments with minimal human demonstration data. In addition, early developer previews of NVIDIA Isaac Sim 5.0 and Isaac Lab 2.2 — open-source robot simulation and learning frameworks optimized for NVIDIA RTX PRO 6000 workstations — are now available on GitHub. Image courtesy of Wandelbots. Robotics Leaders Tap NVIDIA Simulation Technology to Develop and Deploy Humanoids and More  Robotics developers and solutions providers across the globe are integrating NVIDIA’s three computers to train, simulate and deploy robots. NEURA Robotics, a German robotics company and pioneer for cognitive robots, unveiled the third generation of its humanoid, 4NE1, designed to assist humans in domestic and professional environments through advanced cognitive capabilities and humanlike interaction. 4NE1 is powered by GR00T N1 and was trained in Isaac Sim and Isaac Lab before real-world deployment. NEURA Robotics is also presenting Neuraverse, a digital twin and interconnected ecosystem for robot training, skills and applications, fully compatible with NVIDIA Omniverse technologies. Delta Electronics, a global leader in power management and smart green solutions, is debuting two next-generation collaborative robots: D-Bot Mar and D-Bot 2 in 1 — both trained using Omniverse and Isaac Sim technologies and libraries. These cobots are engineered to transform intralogistics and optimize production flows. Wandelbots, the creator of the Wandelbots NOVA software platform for industrial robotics, is partnering with SoftServe, a global IT consulting and digital services provider, to scale simulation-first automating using NVIDIA Isaac Sim, enabling virtual validation and real-world deployment with maximum impact. Cyngn, a pioneer in autonomous mobile robotics, is integrating its DriveMod technology into Isaac Sim to enable large-scale, high fidelity virtual testing of advanced autonomous operation. Purpose-built for industrial applications, DriveMod is already deployed on vehicles such as the Motrec MT-160 Tugger and BYD Forklift, delivering sophisticated automation to material handling operations. Doosan Robotics, a company specializing in AI robotic solutions, will showcase its “sim to real” solution, using NVIDIA Isaac Sim and cuRobo. Doosan will be showcasing how to seamlessly transfer tasks from simulation to real robots across a wide range of applications — from manufacturing to service industries. Franka Robotics has integrated Isaac GR00T N1.5 into a dual-arm Franka Research 3 (FR3) robot for robotic control. The integration of GR00T N1.5 allows the system to interpret visual input, understand task context and autonomously perform complex manipulation — without the need for task-specific programming or hardcoded logic. Image courtesy of Franka Robotics. Hexagon, the global leader in measurement technologies, launched its new humanoid, dubbed AEON. With its unique locomotion system and multimodal sensor fusion, and powered by NVIDIA’s three-computer solution, AEON is engineered to perform a wide range of industrial applications, from manipulation and asset inspection to reality capture and operator support. Intrinsic, a software and AI robotics company, is integrating Intrinsic Flowstate with  Omniverse and OpenUSD for advanced visualization and digital twins that can be used in many industrial use cases. The company is also using NVIDIA foundation models to enhance robot capabilities like grasp planning through AI and simulation technologies. SCHUNK, a global leader in gripping systems and automation technology, is showcasing its innovative grasping kit powered by the NVIDIA Jetson AGX Orin module. The kit intelligently detects objects and calculates optimal grasping points. Schunk is also demonstrating seamless simulation-to-reality transfer using IGS Virtuous software — built on Omniverse technologies — to control a real robot through simulation in a pick-and-place scenario. Universal Robots is showcasing UR15, its fastest cobot yet. Powered by the UR AI Accelerator — developed with NVIDIA and running on Jetson AGX Orin using CUDA-accelerated Isaac libraries — UR15 helps set a new standard for industrial automation. Vention, a full-stack software and hardware automation company, launched its Machine Motion AI, built on CUDA-accelerated Isaac libraries and powered by Jetson. Vention is also expanding its lineup of robotic offerings by adding the FR3 robot from Franka Robotics to its ecosystem, enhancing its solutions for academic and research applications. Image courtesy of Vention. Learn more about the latest robotics advancements by joining NVIDIA at Automatica, running through Friday, June 27. 

Editor’s note: This blog is a part of Into the Omniverse, a series focused on how developers, 3D practitioners and enterprises can transform their workflows using the latest advances in OpenUSD and NVIDIA Omniverse. Simulated driving environments enable engineers to safely and efficiently train, test and validate autonomous vehicles (AVs) across countless real-world and edge-case scenarios without the risks and costs of physical testing. These simulated environments can be created through neural reconstruction of real-world data from AV fleets or generated with world foundation models (WFMs) — neural networks that understand physics and real-world properties. WFMs can be used to generate synthetic datasets for enhanced AV simulation. To help physical AI developers build such simulated environments, NVIDIA unveiled major advances in WFMs at the GTC Paris and CVPR conferences earlier this month. These new capabilities enhance NVIDIA Cosmos — a platform of generative WFMs, advanced tokenizers, guardrails and accelerated data processing tools. Key innovations like Cosmos Predict-2, the Cosmos Transfer-1 NVIDIA preview NIM microservice and Cosmos Reason are improving how AV developers generate synthetic data, build realistic simulated environments and validate safety systems at unprecedented scale. Universal Scene Description (OpenUSD), a unified data framework and standard for physical AI applications, enables seamless integration and interoperability of simulation assets across the development pipeline. OpenUSD standardization plays a critical role in ensuring 3D pipelines are built to scale. NVIDIA Omniverse, a platform of application programming interfaces, software development kits and services for building OpenUSD-based physical AI applications, enables simulations from WFMs and neural reconstruction at world scale. Leading AV organizations — including Foretellix, Mcity, Oxa, Parallel Domain, Plus AI and Uber — are among the first to adopt Cosmos models. Foundations for Scalable, Realistic Simulation Cosmos Predict-2, NVIDIA’s latest WFM, generates high-quality synthetic data by predicting future world states from multimodal inputs like text, images and video. This capability is critical for creating temporally consistent, realistic scenarios that accelerate training and validation of AVs and robots. In addition, Cosmos Transfer, a control model that adds variations in weather, lighting and terrain to existing scenarios, will soon be available to 150,000 developers on CARLA, a leading open-source AV simulator. This greatly expands the broad AV developer community’s access to advanced AI-powered simulation tools. Developers can start integrating synthetic data into their own pipelines using the NVIDIA Physical AI Dataset. The latest release includes 40,000 clips generated using Cosmos. Building on these foundations, the Omniverse Blueprint for AV simulation provides a standardized, API-driven workflow for constructing rich digital twins, replaying real-world sensor data and generating new ground-truth data for closed-loop testing. The blueprint taps into OpenUSD’s layer-stacking and composition arcs, which enable developers to collaborate asynchronously and modify scenes nondestructively. This helps create modular, reusable scenario variants to efficiently generate different weather conditions, traffic patterns and edge cases. Driving the Future of AV Safety To bolster the operational safety of AV systems, NVIDIA earlier this year introduced NVIDIA Halos — a comprehensive safety platform that integrates the company’s full automotive hardware and software stack with AI research focused on AV safety. The new Cosmos models — Cosmos Predict- 2, Cosmos Transfer- 1 NIM and Cosmos Reason — deliver further safety enhancements to the Halos platform, enabling developers to create diverse, controllable and realistic scenarios for training and validating AV systems. These models, trained on massive multimodal datasets including driving data, amplify the breadth and depth of simulation, allowing for robust scenario coverage — including rare and safety-critical events — while supporting post-training customization for specialized AV tasks. At CVPR, NVIDIA was recognized as an Autonomous Grand Challenge winner, highlighting its leadership in advancing end-to-end AV workflows. The challenge used OpenUSD’s robust metadata and interoperability to simulate sensor inputs and vehicle trajectories in semi-reactive environments, achieving state-of-the-art results in safety and compliance. Learn more about how developers are leveraging tools like CARLA, Cosmos, and Omniverse to advance AV simulation in this livestream replay: Hear NVIDIA Director of Autonomous Vehicle Research Marco Pavone on the NVIDIA AI Podcast share how digital twins and high-fidelity simulation are improving vehicle testing, accelerating development and reducing real-world risks. Get Plugged Into the World of OpenUSD Learn more about what’s next for AV simulation with OpenUSD by watching the replay of NVIDIA founder and CEO Jensen Huang’s GTC Paris keynote. Looking for more live opportunities to learn more about OpenUSD? Don’t miss sessions and labs happening at SIGGRAPH 2025, August 10–14. Discover why developers and 3D practitioners are using OpenUSD and learn how to optimize 3D workflows with the self-paced “Learn OpenUSD” curriculum for 3D developers and practitioners, available for free through the NVIDIA Deep Learning Institute. Explore the Alliance for OpenUSD forum and the AOUSD website. Stay up to date by subscribing to NVIDIA Omniverse news, joining the community and following NVIDIA Omniverse on Instagram, LinkedIn, Medium and X.

By TREVOR HOGG Images courtesy of Prime Video. For those seeking an alternative to the MCU, Prime Video has two offerings of the live-action and animated variety that take the superhero genre into R-rated territory where the hands of the god-like figures get dirty, bloodied and severed. “The Boys is about the intersection of celebrity and politics using superheroes,” states Stephan Fleet, VFX Supervisor on The Boys. “Sometimes I see the news and I don’t even know we can write to catch up to it! But we try. Invincible is an intense look at an alternate DC Universe that has more grit to the superhero side of it all. On one hand, I was jealous watching Season 1 of Invincible because in animation you can do things that you can’t do in real life on a budget.” Season 4 does not tone down the blood, gore and body count. Fleet notes, “The writers almost have this dialogue with us. Sometimes, they’ll write in the script, ‘And Fleet will come up with a cool visual effect for how to kill this person.’ Or, ‘Chhiu, our fight coordinator, will make an awesome fight.’ It is a frequent topic of conversation. We’re constantly trying to be inventive and create new ways to kill people!” When Splinter (Rob Benedict) splits in two, the cloning effect was inspired by cellular mitosis. “The writers almost have this dialogue with us. Sometimes, they’ll write in the script, ‘And Fleet will come up with a cool visual effect for how to kill this person.’ Or, ‘Chhiu, our fight coordinator, will make an awesome fight.’ It is a frequent topic of conversation. We’re constantly trying to be inventive and create new ways to kill people!” —Stephan Fleet, VFX Supervisor A total of 1,600 visual effects shots were created for the eight episodes by ILM, Pixomondo, MPC Toronto, Spin VFX, DNEG, Untold Studios, Luma Pictures and Rocket Science VFX. Previs was a critical part of the process. “We have John Griffith [Previs Director], who owns a small company called CNCPT out of Texas, and he does wonderful Unreal Engine level previs,” Fleet remarks. “On set, we have a cartoon of what is going to be done, and you’ll be amazed, specifically for action and heavy visual effects stuff, how close those shots are to the previs when we finish.” Founding Director of Federal Bureau of Superhuman Affairs, Victoria Neuman, literally gets ripped in half by two tendrils coming out of Compound V-enhanced Billy Butcher, the leader of superhero resistance group The Boys. “The word that we like to use on this show is ‘grounded,’ and I like to say ‘grounded’ with an asterisk in this day and age because we’re grounded until we get to killing people in the craziest ways. In this case, having someone floating in the air and being ripped in half by two tendrils was all CG.” Multiple plates were shot to enable Simon Pegg to phase through the actor laying in a hospital bed. Testing can get rather elaborate. “For that end scene with Butcher’s tendrils, the room was two stories, and we were able to put the camera up high along with a bunch of blood cannons,” Fleet recalls. “When the body rips in half and explodes, there is a practical component. We rained down a bunch of real blood and guts right in front of Huey. It’s a known joke that we like to douse Jack Quaid with blood as much as possible! In this case, the special effects team led by Hudson Kenny needed to test it the day before, and I said, “I’ll be the guinea pig for the test.’ They covered the whole place with plastic like it was a Dexter kill room because you don’t want to destroy the set. I’m standing there in a white hazmat suit with goggles on, covered from head to toe in plastic and waiting as they’re tweaking all of these things. It sounds like World War II going on. They’re on walkie talkies to each other, and then all of a sudden, it’s ‘Five, four, three, two, one…’  And I get exploded with blood. I wanted to see what it was like, and it’s intense.” “On set, we have a cartoon of what is going to be done, and you’ll be amazed, specifically for action and heavy visual effects stuff, how close those shots are to the previs when we finish.” —Stephan Fleet, VFX Supervisor The Deep has a love affair with an octopus called Ambrosius, voiced by Tilda Swinton. “It’s implied bestiality!” Fleet laughs. “I would call it more of a romance. What was fun from my perspective is that I knew what the look was going to be [from Season 3], so then it’s about putting in the details and the animation. One of the instincts that you always have when you’re making a sea creature that talks to a human [is] you tend to want to give it human gestures and eyebrows. Erik Kripke [Creator, Executive Producer, Showrunner, Director, Writer] said, ‘No. We have to find things that an octopus could do that conveys the same emotion.’ That’s when ideas came in, such as putting a little The Deep toy inside the water tank. When Ambrosius is trying to have an intimate moment or connect with him, she can wrap a tentacle around that. My favorite experience doing Ambrosius was when The Deep is reading poetry to her on a bed. CG creatures touching humans is one of the more complicated things to do and make look real. Ambrosius’ tentacles reach for his arm, and it becomes an intimate moment. More than touching the skin, displacing the bedsheet as Ambrosius moved ended up becoming a lot of CG, and we had to go back and forth a few times to get that looking right; that turned out to be tricky.” A building is replaced by a massive crowd attending a rally being held by Homelander. In a twisted form of sexual foreplay, Sister Sage has The Deep perform a transorbital lobotomy on her. “Thank you, Amazon for selling lobotomy tools as novelty items!” Fleet chuckles. “We filmed it with a lobotomy tool on set. There is a lot of safety involved in doing something like that. Obviously, you don’t want to put any performer in any situation where they come close to putting anything real near their eye. We created this half lobotomy tool and did this complicated split screen with the lobotomy tool on a teeter totter. The Deep was [acting in a certain way] in one shot and Sister Sage reacted in the other shot. To marry the two ended up being a lot of CG work. Then there are these close-ups which are full CG. I always keep a dummy head that is painted gray that I use all of the time for reference. In macrophotography I filmed this lobotomy tool going right into the eye area. I did that because the tool is chrome, so it’s reflective and has ridges. It has an interesting reflective property. I was able to see how and what part of the human eye reflects onto the tool. A lot of that shot became about realistic reflections and lighting on the tool. Then heavy CG for displacing the eye and pushing the lobotomy tool into it. That was one of the more complicated sequences that we had to achieve.” In order to create an intimate moment between Ambrosius and The Deep, a toy version of the superhero was placed inside of the water tank that she could wrap a tentacle around. “The word that we like to use on this show is ‘grounded,’ and I like to say ‘grounded’ with an asterisk in this day and age because we’re grounded until we get to killing people in the craziest ways. In this case, having someone floating in the air and being ripped in half by two tendrils was all CG.” —Stephan Fleet, VFX Supervisor Sheep and chickens embark on a violent rampage courtesy of Compound V with the latter piercing the chest of a bodyguard belonging to Victoria Neuman. “Weirdly, that was one of our more traditional shots,’ Fleet states. “What is fun about that one is I asked for real chickens as reference. The chicken flying through his chest is real. It’s our chicken wrangler in green suit gently tossing a chicken. We blended two real plates together with some CG in the middle.” A connection was made with a sci-fi classic. “The sheep kill this bull, and we shot it is in this narrow corridor of fencing. When they run, I always equated it as the Trench Run in Star Wars and looked at the sheep as TIE fighters or X-wings coming at them.” The scene was one of the scarier moments for the visual effects team. Fleet explains, “When I read the script, I thought this could be the moment where we jump the shark. For the shots where the sheep are still and scream to the camera, Untold Studios did a bunch of R&D and came up with baboon teeth. I tried to keep anything real as much as possible, but, obviously, when sheep are flying, they have to be CG. I call it the Battlestar Galactica theory, where I like to shake the camera, overshoot shots and make it sloppy when they’re in the air so you can add motion blur. Comedy also helps sell visual effects.” The sheep injected with Compound V develop the ability to fly and were shot in an imperfect manner to help ground the scenes. Once injected with Compound V, Hugh Campbell Sr. (Simon Pegg) develops the ability to phase through objects, including human beings. “We called it the Bro-nut because his name in the script is Wall Street Bro,” Fleet notes. “That was a complicated motion control shot, repeating the move over and over again. We had to shoot multiple plates of Simon Pegg and the guy in the bed. Special effects and prosthetics created a dummy guy with a hole in his chest with practical blood dripping down. It was meshing it together and getting the timing right in post. On top of that, there was the CG blood immediately around Simon Pegg.” The phasing effect had to avoid appearing as a dissolve. “I had this idea of doing high-frequency vibration on the X axis loosely based on how The Flash vibrates through walls. You want everything to have a loose motivation that then helps trigger the visuals. We tried not to overcomplicate that because, ultimately, you want something like that to be quick. If you spend too much time on phasing, it can look cheesy. In our case, it was a lot of false walls. Simon Pegg is running into a greenscreen hole which we plug in with a wall or coming out of one. I went off the actor’s action, and we added a light opacity mix with some X-axis shake.” Providing a different twist to the fights was the replacement of spurting blood with photoreal rubber duckies during a drug-induced hallucination. Homelander (Anthony Starr) breaks a mirror which emphasizes his multiple personality disorder. “The original plan was that special effects was going to pre-break a mirror, and we were going to shoot Anthony Starr moving his head doing all of the performances in the different parts of the mirror,” Fleet reveals. “This was all based on a photo that my ex-brother-in-law sent me. He was walking down a street in Glendale, California, came across a broken mirror that someone had thrown out, and took a photo of himself where he had five heads in the mirror. We get there on the day, and I’m realizing that this is really complicated. Anthony has to do these five different performances, and we have to deal with infinite mirrors. At the last minute, I said, ‘We have to do this on a clean mirror.’ We did it on a clear mirror and gave Anthony different eyelines. The mirror break was all done in post, and we were able to cheat his head slightly and art-direct where the break crosses his chin. Editorial was able to do split screens for the timing of the dialogue.” “For the shots where the sheep are still and scream to the camera, Untold Studios did a bunch of R&D and came up with baboon teeth. I tried to keep anything real as much as possible, but, obviously, when sheep are flying, they have to be CG. I call it the Battlestar Galactica theory, where I like to shake the camera, overshoot shots and make it sloppy when they’re in the air so you can add motion blur. Comedy also helps sell visual effects.” —Stephan Fleet, VFX Supervisor Initially, the plan was to use a practical mirror, but creating a digital version proved to be the more effective solution. A different spin on the bloodbath occurs during a fight when a drugged Frenchie (Tomer Capone) hallucinates as Kimiko Miyashiro (Karen Fukuhara) goes on a killing spree. “We went back and forth with a lot of different concepts for what this hallucination would be,” Fleet remarks. “When we filmed it, we landed on Frenchie having a synesthesia moment where he’s seeing a lot of abstract colors flying in the air. We started getting into that in post and it wasn’t working. We went back to the rubber duckies, which goes back to the story of him in the bathtub. What’s in the bathtub? Rubber duckies, bubbles and water. There was a lot of physics and logic required to figure out how these rubber duckies could float out of someone’s neck. We decided on bubbles when Kimiko hits people’s heads. At one point, we had water when she got shot, but it wasn’t working, so we killed it. We probably did about 100 different versions. We got really detailed with our rubber duckie modeling because we didn’t want it to look cartoony. That took a long time.” Ambrosius, voiced by Tilda Swinton, gets a lot more screentime in Season 4. When Splinter (Rob Benedict) splits in two was achieved heavily in CG. “Erik threw out the words ‘cellular mitosis’ early on as something he wanted to use,” Fleet states. “We shot Rob Benedict on a greenscreen doing all of the different performances for the clones that pop out. It was a crazy amount of CG work with Houdini and particle and skin effects. We previs’d the sequence so we had specific actions. One clone comes out to the right and the other pulls backwards.” What tends to go unnoticed by many is Splinter’s clones setting up for a press conference being held by Firecracker (Valorie Curry). “It’s funny how no one brings up the 22-hour motion control shot that we had to do with Splinter on the stage, which was the most complicated shot!” Fleet observes. “We have this sweeping long shot that brings you into the room and follows Splinter as he carries a container to the stage and hands it off to a clone, and then you reveal five more of them interweaving each other and interacting with all of these objects. It’s like a minute-long dance. First off, you have to choreograph it. We previs’d it, but then you need to get people to do it. We hired dancers and put different colored armbands on them. The camera is like another performer, and a metronome is going, which enables you to find a pace. That took about eight hours of rehearsal. Then Rob has to watch each one of their performances and mimic it to the beat. When he is handing off a box of cables, it’s to a double who is going to have to be erased and be him on the other side. They have to be almost perfect in their timing and lineup in order to take it over in visual effects and make it work.”

Mark Theriault founded the startup FITY envisioning a line of clever cooling products: cold drink holders that come with freezable pucks to keep beverages cold for longer without the mess of ice. The entrepreneur started with 3D prints of products in his basement, building one unit at a time, before eventually scaling to mass production. Founding a consumer product company from scratch was a tall order for a single person. Going from preliminary sketches to production-ready designs was a major challenge. To bring his creative vision to life, Theriault relied on AI and his NVIDIA GeForce RTX-equipped system. For him, AI isn’t just a tool — it’s an entire pipeline to help him accomplish his goals. Read more about his workflow below. Plus, GeForce RTX 5050 laptops start arriving today at retailers worldwide, from $999. GeForce RTX 5050 Laptop GPUs feature 2,560 NVIDIA Blackwell CUDA cores, fifth-generation AI Tensor Cores, fourth-generation RT Cores, a ninth-generation NVENC encoder and a sixth-generation NVDEC decoder. In addition, NVIDIA’s Plug and Play: Project G-Assist Plug-In Hackathon — running virtually through Wednesday, July 16 — invites developers to explore AI and build custom G-Assist plug-ins for a chance to win prizes. Save the date for the G-Assist Plug-In webinar on Wednesday, July 9, from 10-11 a.m. PT, to learn more about Project G-Assist capabilities and fundamentals, and to participate in a live Q&A session. From Concept to Completion To create his standout products, Theriault tinkers with potential FITY Flex cooler designs with traditional methods, from sketch to computer-aided design to rapid prototyping, until he finds the right vision. A unique aspect of the FITY Flex design is that it can be customized with fun, popular shoe charms. For packaging design inspiration, Theriault uses his preferred text-to-image generative AI model for prototyping, Stable Diffusion XL — which runs 60% faster with the NVIDIA TensorRT software development kit — using the modular, node-based interface ComfyUI. ComfyUI gives users granular control over every step of the generation process — prompting, sampling, model loading, image conditioning and post-processing. It’s ideal for advanced users like Theriault who want to customize how images are generated. Theriault’s uses of AI result in a complete computer graphics-based ad campaign. Image courtesy of FITY. NVIDIA and GeForce RTX GPUs based on the NVIDIA Blackwell architecture include fifth-generation Tensor Cores designed to accelerate AI and deep learning workloads. These GPUs work with CUDA optimizations in PyTorch to seamlessly accelerate ComfyUI, reducing generation time on FLUX.1-dev, an image generation model from Black Forest Labs, from two minutes per image on the Mac M3 Ultra to about four seconds on the GeForce RTX 5090 desktop GPU. ComfyUI can also add ControlNets — AI models that help control image generation — that Theriault uses for tasks like guiding human poses, setting compositions via depth mapping and converting scribbles to images. Theriault even creates his own fine-tuned models to keep his style consistent. He used low-rank adaptation (LoRA) models — small, efficient adapters into specific layers of the network — enabling hyper-customized generation with minimal compute cost. LoRA models allow Theriault to ideate on visuals quickly. Image courtesy of FITY. “Over the last few months, I’ve been shifting from AI-assisted computer graphics renders to fully AI-generated product imagery using a custom Flux LoRA I trained in house. My RTX 4080 SUPER GPU has been essential for getting the performance I need to train and iterate quickly.” – Mark Theriault, founder of FITY  Theriault also taps into generative AI to create marketing assets like FITY Flex product packaging. He uses FLUX.1, which excels at generating legible text within images, addressing a common challenge in text-to-image models. Though FLUX.1 models can typically consume over 23GB of VRAM, NVIDIA has collaborated with Black Forest Labs to help reduce the size of these models using quantization — a technique that reduces model size while maintaining quality. The models were then accelerated with TensorRT, which provides an up to 2x speedup over PyTorch. To simplify using these models in ComfyUI, NVIDIA created the FLUX.1 NIM microservice, a containerized version of FLUX.1 that can be loaded in ComfyUI and enables FP4 quantization and TensorRT support. Combined, the models come down to just over 11GB of VRAM, and performance improves by 2.5x. Theriault uses the Blender Cycles app to render out final files. For 3D workflows, NVIDIA offers the AI Blueprint for 3D-guided generative AI to ease the positioning and composition of 3D images, so anyone interested in this method can quickly get started. Photorealistic renders. Image courtesy of FITY. Finally, Theriault uses large language models to generate marketing copy — tailored for search engine optimization, tone and storytelling — as well as to complete his patent and provisional applications, work that usually costs thousands of dollars in legal fees and considerable time. Generative AI helps Theriault create promotional materials like the above. Image courtesy of FITY. “As a one-man band with a ton of content to generate, having on-the-fly generation capabilities for my product designs really helps speed things up.” – Mark Theriault, founder of FITY Every texture, every word, every photo, every accessory was a micro-decision, Theriault said. AI helped him survive the “death by a thousand cuts” that can stall solo startup founders, he added. Each week, the RTX AI Garage blog series features community-driven AI innovations and content for those looking to learn more about NVIDIA NIM microservices and AI Blueprints, as well as building AI agents, creative workflows, digital humans, productivity apps and more on AI PCs and workstations.  Plug in to NVIDIA AI PC on Facebook, Instagram, TikTok and X — and stay informed by subscribing to the RTX AI PC newsletter. Follow NVIDIA Workstation on LinkedIn and X.  See notice regarding software product information.

By TREVOR HOGG Images courtesy of Warner Bros. Pictures. Rather than a world constructed around photorealistic pixels, a video game created by Markus Persson has taken the boxier 3D voxel route, which has become its signature aesthetic, and sparked an international phenomenon that finally gets adapted into a feature with the release of A Minecraft Movie. Brought onboard to help filmmaker Jared Hess in creating the environments that the cast of Jason Momoa, Jack Black, Sebastian Hansen, Emma Myers and Danielle Brooks find themselves inhabiting was Disguise under the direction of Production VFX Supervisor Dan Lemmon. “[A]s the Senior Unreal Artist within the Virtual Art Department (VAD) on Minecraft, I experienced the full creative workflow. What stood out most was how deeply the VAD was embedded across every stage of production. We weren’t working in isolation. From the production designer and director to the VFX supervisor and DP, the VAD became a hub for collaboration.” —Talia Finlayson, Creative Technologist, Disguise Interior and exterior environments had to be created, such as the shop owned by Steve (Jack Black). “Prior to working on A Minecraft Movie, I held more technical roles, like serving as the Virtual Production LED Volume Operator on a project for Apple TV+ and Paramount Pictures,” notes Talia Finlayson, Creative Technologist for Disguise. “But as the Senior Unreal Artist within the Virtual Art Department (VAD) on Minecraft, I experienced the full creative workflow. What stood out most was how deeply the VAD was embedded across every stage of production. We weren’t working in isolation. From the production designer and director to the VFX supervisor and DP, the VAD became a hub for collaboration.” The project provided new opportunities. “I’ve always loved the physicality of working with an LED volume, both for the immersion it provides and the way that seeing the environment helps shape an actor’s performance,” notes Laura Bell, Creative Technologist for Disguise. “But for A Minecraft Movie, we used Simulcam instead, and it was an incredible experience to live-composite an entire Minecraft world in real-time, especially with nothing on set but blue curtains.” Set designs originally created by the art department in Rhinoceros 3D were transformed into fully navigable 3D environments within Unreal Engine. “These scenes were far more than visualizations,” Finlayson remarks. “They were interactive tools used throughout the production pipeline. We would ingest 3D models and concept art, clean and optimize geometry using tools like Blender, Cinema 4D or Maya, then build out the world in Unreal Engine. This included applying materials, lighting and extending environments. These Unreal scenes we created were vital tools across the production and were used for a variety of purposes such as enabling the director to explore shot compositions, block scenes and experiment with camera movement in a virtual space, as well as passing along Unreal Engine scenes to the visual effects vendors so they could align their digital environments and set extensions with the approved production layouts.” A virtual exploration of Steve’s shop in Midport Village. Certain elements have to be kept in mind when constructing virtual environments. “When building virtual environments, you need to consider what can actually be built, how actors and cameras will move through the space, and what’s safe and practical on set,” Bell observes. “Outside the areas where strict accuracy is required, you want the environments to blend naturally with the original designs from the art department and support the story, creating a space that feels right for the scene, guides the audience’s eye and sets the right tone. Things like composition, lighting and small environmental details can be really fun to work on, but also serve as beautiful additions to help enrich a story.” “I’ve always loved the physicality of working with an LED volume, both for the immersion it provides and the way that seeing the environment helps shape an actor’s performance. But for A Minecraft Movie, we used Simulcam instead, and it was an incredible experience to live-composite an entire Minecraft world in real-time, especially with nothing on set but blue curtains.” —Laura Bell, Creative Technologist, Disguise Among the buildings that had to be created for Midport Village was Steve’s (Jack Black) Lava Chicken Shack. Concept art was provided that served as visual touchstones. “We received concept art provided by the amazing team of concept artists,” Finlayson states. “Not only did they send us 2D artwork, but they often shared the 3D models they used to create those visuals. These models were incredibly helpful as starting points when building out the virtual environments in Unreal Engine; they gave us a clear sense of composition and design intent. Storyboards were also a key part of the process and were constantly being updated as the project evolved. Having access to the latest versions allowed us to tailor the virtual environments to match camera angles, story beats and staging. Sometimes we would also help the storyboard artists by sending through images of the Unreal Engine worlds to help them geographically position themselves in the worlds and aid in their storyboarding.” At times, the video game assets came in handy. “Exteriors often involved large-scale landscapes and stylized architectural elements, which had to feel true to the Minecraft world,” Finlayson explains. “In some cases, we brought in geometry from the game itself to help quickly block out areas. For example, we did this for the Elytra Flight Chase sequence, which takes place through a large canyon.” Flexibility was critical. “A key technical challenge we faced was ensuring that the Unreal levels were built in a way that allowed for fast and flexible iteration,” Finlayson remarks. “Since our environments were constantly being reviewed by the director, production designer, DP and VFX supervisor, we needed to be able to respond quickly to feedback, sometimes live during a review session. To support this, we had to keep our scenes modular and well-organized; that meant breaking environments down into manageable components and maintaining clean naming conventions. By setting up the levels this way, we could make layout changes, swap assets or adjust lighting on the fly without breaking the scene or slowing down the process.” Production schedules influence the workflows, pipelines and techniques. “No two projects will ever feel exactly the same,” Bell notes. “For example, Pat Younis [VAD Art Director] adapted his typical VR setup to allow scene reviews using a PS5 controller, which made it much more comfortable and accessible for the director. On a more technical side, because everything was cubes and voxels, my Blender workflow ended up being way heavier on the re-mesh modifier than usual, definitely not something I’ll run into again anytime soon!” A virtual study and final still of the cast members standing outside of the Lava Chicken Shack. “We received concept art provided by the amazing team of concept artists. Not only did they send us 2D artwork, but they often shared the 3D models they used to create those visuals. These models were incredibly helpful as starting points when building out the virtual environments in Unreal Engine; they gave us a clear sense of composition and design intent. Storyboards were also a key part of the process and were constantly being updated as the project evolved. Having access to the latest versions allowed us to tailor the virtual environments to match camera angles, story beats and staging.” —Talia Finlayson, Creative Technologist, Disguise The design and composition of virtual environments tended to remain consistent throughout principal photography. “The only major design change I can recall was the removal of a second story from a building in Midport Village to allow the camera crane to get a clear shot of the chicken perched above Steve’s lava chicken shack,” Finlayson remarks. “I would agree that Midport Village likely went through the most iterations,” Bell responds. “The archway, in particular, became a visual anchor across different levels. We often placed it off in the distance to help orient both ourselves and the audience and show how far the characters had traveled. I remember rebuilding the stairs leading up to the rampart five or six times, using different configurations based on the physically constructed stairs. This was because there were storyboarded sequences of the film’s characters, Henry, Steve and Garrett, being chased by piglins, and the action needed to match what could be achieved practically on set.” Virtually conceptualizing the layout of Midport Village. Complex virtual environments were constructed for the final battle and the various forest scenes throughout the movie. “What made these particularly challenging was the way physical set pieces were repurposed and repositioned to serve multiple scenes and locations within the story,” Finlayson reveals. “The same built elements had to appear in different parts of the world, so we had to carefully adjust the virtual environments to accommodate those different positions.” Bell is in agreement with her colleague. “The forest scenes were some of the more complex environments to manage. It could get tricky, particularly when the filming schedule shifted. There was one day on set where the order of shots changed unexpectedly, and because the physical sets looked so similar, I initially loaded a different perspective than planned. Fortunately, thanks to our workflow, Lindsay George [VP Tech] and I were able to quickly open the recorded sequence in Unreal Engine and swap out the correct virtual environment for the live composite without any disruption to the shoot.” An example of the virtual and final version of the Woodland Mansion. “Midport Village likely went through the most iterations. The archway, in particular, became a visual anchor across different levels. We often placed it off in the distance to help orient both ourselves and the audience and show how far the characters had traveled.” —Laura Bell, Creative Technologist, Disguise Extensive detail was given to the center of the sets where the main action unfolds. “For these areas, we received prop layouts from the prop department to ensure accurate placement and alignment with the physical builds,” Finlayson explains. “These central environments were used heavily for storyboarding, blocking and department reviews, so precision was essential. As we moved further out from the practical set, the environments became more about blocking and spatial context rather than fine detail. We worked closely with Production Designer Grant Major to get approval on these extended environments, making sure they aligned with the overall visual direction. We also used creatures and crowd stand-ins provided by the visual effects team. These gave a great sense of scale and placement during early planning stages and allowed other departments to better understand how these elements would be integrated into the scenes.” Cast members Sebastian Hansen, Danielle Brooks and Emma Myers stand in front of the Earth Portal Plateau environment. Doing a virtual scale study of the Mountainside. Practical requirements like camera moves, stunt choreography and crane setups had an impact on the creation of virtual environments. “Sometimes we would adjust layouts slightly to open up areas for tracking shots or rework spaces to accommodate key action beats, all while keeping the environment feeling cohesive and true to the Minecraft world,” Bell states. “Simulcam bridged the physical and virtual worlds on set, overlaying Unreal Engine environments onto live-action scenes in real-time, giving the director, DP and other department heads a fully-realized preview of shots and enabling precise, informed decisions during production. It also recorded critical production data like camera movement paths, which was handed over to the post-production team to give them the exact tracks they needed, streamlining the visual effects pipeline.” Piglots cause mayhem during the Wingsuit Chase. Virtual versions of the exterior and interior of the Safe House located in the Enchanted Woods. “One of the biggest challenges for me was managing constant iteration while keeping our environments clean, organized and easy to update,” Finlayson notes. “Because the virtual sets were reviewed regularly by the director and other heads of departments, feedback was often implemented live in the room. This meant the environments had to be flexible. But overall, this was an amazing project to work on, and I am so grateful for the incredible VAD team I was a part of – Heide Nichols [VAD Supervisor], Pat Younis, Jake Tuck [Unreal Artist] and Laura. Everyone on this team worked so collaboratively, seamlessly and in such a supportive way that I never felt like I was out of my depth.” There was another challenge that is more to do with familiarity. “Having a VAD on a film is still a relatively new process in production,” Bell states. “There were moments where other departments were still learning what we did and how to best work with us. That said, the response was overwhelmingly positive. I remember being on set at the Simulcam station and seeing how excited people were to look at the virtual environments as they walked by, often stopping for a chat and a virtual tour. Instead of seeing just a huge blue curtain, they were stoked to see something Minecraft and could get a better sense of what they were actually shooting.”