"The model was never the hard part."
We’re introducing Physical AI Workbench for everything in physical AI that isn’t the model: @nvidia Cosmos 3, Isaac Lab, GR00T, and other physical AI tools, pre-wired, agent-driven, open source.
https://t.co/YGiLnaoIec
NVIDIA Nemotron™ 3 Ultra is now live on Nebius Token Factory.
It’s built for long-running agents across coding, deep research and enterprise workflows. Nemotron 3 Ultra delivers frontier reasoning with up to 5x faster inference and up to 30% lower cost for agentic workloads.
For builders, the next question is production: performance, reliability, economics and control.
Run it today: https://t.co/Jjm8SfLRRH
NVIDIA Nemotron™ 3 Ultra is now live on Nebius Token Factory.
It’s built for long-running agents across coding, deep research and enterprise workflows.
Nemotron 3 Ultra delivers frontier reasoning with up to 5x faster inference and up to 30% lower cost for agentic workloads.
For builders, the next question is production: performance, reliability, economics and control.
Run it today: https://t.co/kHWvYquwoy
@NVIDIA's Build Your Own Transaction Foundation Model developer example is now available with 1-click deployment on Nebius AI Cloud.
@Revolut App already runs PRAGMA in production this way: 40B events, 200+ H100 GPUs, 3x faster pre-training.
Discover more -> https://t.co/aiIViLtdCx
Nebius choisit la France.
L’acteur du Cloud IA dévoile un projet de plus de 8 milliards d’euros d’investissement pour déployer des infrastructures et des services de cloud, avec une capacité cible de 240 MW, positionnant le site parmi les plus puissants du continent.
Bedankt!
BuilderShip is an AI hackathon co-hosted by @nebiusai, @composio, @tavilyai, and @openclaw. Finals are on a yacht on June 14, San Francisco Bay.
To apply: post something you've built (agent, demo, or repo) and tag all four accounts.
Every builder gets GPU credits, Token Factory inference keys, Composio integrations, Tavily search, and OpenClaw runtime from day one.
Top 30 builders make the finals. Winner takes home $50K in cloud credits and a DGX Spark.
Submit by June 12 → https://t.co/OHzftpPcva
Congrats to the JetBrains team on the Mellum2 launch! Nebius was part of the training infrastructure for this model. Excited to see what teams build with it.
Mellum started with code completion.
Mellum2 is built for more – handling both natural language and code.
A 12B-parameter open-source LLM for routing, RAG, and sub-agents, optimized for ultra-low-latency inference.
Now on @huggingface.
Learn more: https://t.co/28sG8Ql52L
Our customer @reactorworld just came out of stealth, and if you're building with world models or want to get started with them, this is big news. They've created exactly what developers need for the world model era: low latency, global scale, real-time setup with minimal effort. The team is exceptional (Apple, Meta, Google, Luma AI alumni) and the backers are serious ($59M, Lightspeed-led). And they're building on Nebius. Worth following!
Today Reactor is coming out of stealth. We’ve raised $59M in Seed and Series A funding, led by @lightspeedvp, with participation from @AmplifyPartners, @wndrco, @Sky9Capital, and @FPVventures.
Reactor is the platform for building in the World Model era: the infrastructure that lets developers build with them at global scale for the first time. Stream from a frontier World Model to your app, in real time, all in under 10 lines of code.
World Models represent the next major shift in AI: pixels, audio and actions are generated on the fly, in real-time, in response to user inputs, and to the environment. Every time computing has made a shift from passive to interactive, entire industries appeared that didn't exist before. We're standing in front of such moment again.
Over the last 6 months, we’ve assembled an all-star team with alumni from Apple, Meta, Google, Luma AI, Netflix, and Replicate. We're already partnering with some of the biggest names and labs in the world, and hundreds of developers are already building on Reactor.
The World Model era starts now.
We're building one of Europe's largest AI factories in Lappeenranta, Finland, and we're looking for the team to run it.
Latest-generation NVIDIA accelerated compute. First capacity live in 2027, powered by low-carbon energy.
From shift patterns to tooling, this is a chance to write the runbook from day one across data center operations and data center IT.
Bring it online with us: https://t.co/ojPiuxzG9Y
What does it actually take to move AI from pilot to production in digital health?
On 16 June, Nebius is hosting a conversation for the people navigating this question: senior builders and decision-makers in healthcare and life sciences who are moving past the pilot stage and into real-world AI deployments in healthcare.
16 June | 17:30–21:00 CET | Nebius Amsterdam
Apply here: https://t.co/Obj94lrDjM
Training an #AI world model is demanding. A single cluster failure means massive rework.
After reliability issues across 3 cloud providers, Jua moved to Nebius. Thanks to Nebius AI's stable infrastructure powered by NVIDIA GPUs and high-speed NVIDIA InfiniBand networking, Jua’s EPT-2 now has state-of-the-art performance globally.
Result: EPT-2 runs at 6x higher temporal and spatial resolution than comparable AI models. Read more: https://t.co/WEeNaBF73A
"At university, most problems have one correct answer. At work, problems often have more than one solution, each with trade-offs.”
That's Andrei Strelkovskii, Principal SWE at Nebius, and one of the mentors in our Early Talent Program. He works with the team building storage systems that scale well and respond with sub-millisecond latency.
Interested in the challenge? Applications close May 31 → https://t.co/4JpyKgnVkR
Nebius and @Bloom_Energy have announced a partnership to deploy Bloom’s clean fuel cells to power Nebius’s AI infrastructure build-out.
Nebius selected Bloom for its fast time to power, clean, virtually non-polluting technology, and ability to support the extreme performance demands of AI workloads.
Andrey Korolenko, Chief Product and Infrastructure Officer at Nebius:
“Power remains a key constraint for AI infrastructure build-outs. We chose Bloom because their fuel cells solve that directly: Clean power with virtually no pollutants is deployed onsite, on the timelines our customers need, with the availability AI workloads require.”
Read the full press release: https://t.co/EqxetNNkZr
@Revolut has built one of the fastest growing and most respected financial institutions thanks to their deliberate, multi-year investment in AI powered products and systems.
Over the span of just 3 years, Revolut was able to increase their user base from 38 million to 70 million offering app-based banking and financial services supporting payments, spending, savings, investing, foreign exchange and business accounts.
Read the full story: https://t.co/QPGtTF2v1w
Big week for @recraftai. They just launched V4.1, and the model is now ranked among the top 5 image generation labs in the world on @Designarena. Trained on Nebius. Congrats team!
Nebius Build is coming to London on 21st of May! 🇬🇧
This event is for technical builders working on how AI systems actually get deployed and scaled in the real world. We’ll dive into practical topics like distributed training, post-training, inference, agentic workflows, and running AI systems in production.
Seats are limited. Grab your spot now!
https://t.co/KLFtjGJSIV
Nebius and @LangChain have partnered to integrate Nebius Token Factory with LangChain's Deep Agents. The integration, combined with LangChain's existing Tavily integration, gives teams building on LangChain a direct path to run agent workloads on production-grade AI infrastructure with open-source models, dedicated endpoints, real-time search, and full control over cost and data.
Read the blog to learn more: https://t.co/ifT7GSv5YI