Open source is powerful. We can quickly combine different AI research results, and create beautiful outputs like this. 🔜Coming soon

Everything you need to know about LiveInfra presented by FTK Staking TLDR The LiveInfra works to enhance the Community Arbitrum Node. Creating a free RPC service built and operated for the Livepeer community, by achieving 99.9% uptime through additional compute, improved

Realtime AI video generation and world models are coming together. Watch this space 👀

Earlier this week OpenAI released Sora 2. It's a massive step up in the generated video quality and personalization enabled with their cameo feature, letting you drop yourself into any generated scene via a prompt. The future of media creation is clearly AI assisted, and the

Try the new LongLive video model on your own GPU!

Excited to support StreamDiffusionV2 in the latest release of Scope today! As a reminder, Scope is tool for running and customizing real-time, interactive generative AI pipelines and models. Here is a walkthrough of it in action in Scope (see next post if you just want the

A full day of intense hacking in Lisbon 🇵🇹 What will they BUIDL 👀

I gave a talk at this week's Livepeer summit highlighting the history of innovations that the project has spurred in the decentralized tech space over the last decade, all while powering millions of minutes of video processing per week for global streaming and AI applications:

Next Monday's Realtime AI Video Summit during open source AI week in SF is going to be looked back upon in a couple years as a seminal event early in the maturity of what will be a huge video world model market. The authors of StreamDiffusion, StreamV2V, and Self-Forcing amongst

Livepeer hits all-time-highs in Q3 🚀 Q3 2025 marked an ATH of +76% in fees and +94% in network usage, driven by real-time AI video compute processing +1M minutes per day. Decentralized infrastructure is now powering the next era of AI-driven media. Read Messari’s State of

The Livepeer network has transformed itself into a key GPU infrastructure powering the future of AI video. Last quarter fees rose 76% to an all time high, and now 72% of the fees derive from AI video processing. Check out the full Messari report.

Big props to Helium🎈 for leading the charge on DePIN fee generation. It takes years to lay the infrastructure groundwork, productize, and go to market. DePIN is for the builders with endurance. (slide credit to a16z crypto)

Big new feature release in Scope, allowing you to export, share, and remix your prompt and setting timelines to craft killer AI videos. Check it out.

"What does it mean to play with video generation as replayable and remixable world exploration?" youtube.com/watch?v=yeoGzS…

Introducing KREA AI Realtime 14B, now running inside Daydream Scope. ✨ Type a prompt, watch it instantly come alive as interactive AI video. 📼 Update prompts live. Remix forever. All open-source. ⚡ Compatible RTX 5090 and above.

Who’s coming to Buenos Aires? We’re gathering to talk AI, video, media, and opportunities in the scene Join us for collaborative sessions, with a post-event output of a zine. Aiming for something special. Presented by Livepeer in collab with Refraction | IRL

More real-time AI models...Go Play!

Pack your bags ✈️ 🇦🇷 🥩🟢🎥 Breakout discussions, lightnight talks (no shilling), post-production zine, cheeky happy hour. Presented by Livepeer, curated by Refraction | IRL. Huge shout out to Kaitlyn Davies, Nick Hollins, and lizzy 🌍 for making this a reality 🔥

If you're new to the coming concept of World Models, give this piece by Fei-Fei Li a read. These models will be generative, multimodal, and interactive. They'll support breakthroughs in robotics, science, and many fields of entertainment such as interactive gaming.

Last month, something special happened in the heart of Portugal: builders, core contributors, and ecosystem visionaries came together for two action-packed days dedicated to moving the Livepeer network forward.