NVIDIA Cosmos
AI overview
Sign in and the AI will write an overview from our coverage.
Headlines · 3
- ByteDance world model demo surfaces, reportedly overseen by Zhang Yiming
A Xiaohongshu post circulating in Chinese tech circles says Tim of the filmmaking channel MediaStorm demonstrated ByteDance's latest world model, which could build a world and let users interact with it — freely controlling camera moves, stepping through a virtual 'anywhere door', and letting robots learn manipulation or doctors train by simulation; ByteDance has not announced the model and the details are unverified. Earlier reports said ByteDance is developing a real-time spatial video generation model based on its Seedance video model, personally coordinated by founder Zhang Yiming across research, cloud infrastructure, Pico hardware and content teams with priority computing support; the model generates 20 fps video on demand with roughly 50ms end-to-end latency, rendering in the cloud and streaming to devices, with planned Pico headset support. World models were reportedly listed first among ByteDance's four 2026 AI priorities, with a goal of shipping at least one model benchmarked against Google's Genie 3 by year-end for livestreaming, interactive short dramas, games and simulation for robots and job training.
钛媒体 · 🔥 5 - European cities deploy NVIDIA-powered AI agents for urban operations
An NVIDIA blog post says cities across EMEA are working with local software, infrastructure and integration partners to build physical AI agents and digital twins on NVIDIA accelerated computing for decisions such as road repair, energy use, urban planning and safety. The stack draws on NVIDIA Cosmos open models, the Metropolis video search and summarization (VSS) blueprint, Nemotron open models and Omniverse libraries; in France's Côte-d'Or, Akila and ThinkDeep built operations agents over 500-plus public buildings in a sovereign digital twin, which the post says cut document search from weeks to minutes and reduced operating costs by 10%, targeting 30% at scale. Other cited projects include Galene.AI with Dell Technologies and TD Synnex deploying a local AI factory in the Milan region, a Port of Antwerp-Bruges test using synthetic data for oil-spill warnings, a Citiverse digital twin for Gothenburg, and Honeywell's Forge Cognition at the Dubai Fairmont Hotel; NVIDIA will showcase the work at Smart City Expo in Barcelona on Nov. 3-5.
NVIDIA Blog · 🔥 0 - Standard Bots on its industrial robot AI stack: targeted data, on-prem inference
In a Latent Space interview, Standard Bots co-founder/CEO Evan Beard and head of AI Leif Jentoft described the AI stack behind its industrial robot arms. The company calls itself "America's largest AI-native industrial robot manufacturer," recently raised a $200M Series C at a $1B valuation led by General Catalyst and RoboStrategy, and lists NASA, Amazon and Lockheed Martin as customers. Its largest model sits in the low billions of parameters, trains in the cloud and runs inference on-premises; a zero-shot perception model trained on over a billion images locates and identifies parts, while motion and cell logic stay conventional. Its StandardOS platform offers APIs and SDKs so developers can bring their own models, such as NVIDIA Cosmos.
Latent Space · 🔥 0
Experience and discussion from the community
Share my NVIDIA Cosmos experienceAsk about NVIDIA Cosmos
Nobody has shared their experience with NVIDIA Cosmos yet.