Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

We built a better alternative to json-render & A2UI Up to 3x faster rendering & 67% fewer tokens. Far fewer broken outputs. We've been shipping Generative UI interfaces to 10,000+ developers for the past year. We know where JSON-based rendering breaks in production. So we built a new format... show more

Rabi Shanker Guha

2,334 subscribers

133,987 views • 4 months ago •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

We promised to not change the core experience in Keystone, and we're sticking by that. But we did think it was worthwhile experimenting with a couple of fun, optional, extras now we have a new rendering engine. What do you think?

We promised to not change the core experience in Keystone, and we're sticking by that. But we did think it was worthwhile experimenting with a couple of fun, optional, extras now we have a new rendering engine. What do you think?

Macklebee

10,378 views • 17 days ago

Introducing Kaleido💮 from AI at Meta — a universal generative neural rendering engine for photorealistic, unified object and scene view synthesis. Kaleido is built on a simple but powerful design philosophy: 3D perception is a form of visual common sense. Following this idea, we formulate rendering purely as a sequence-to-sequence generation problem, successfully unifying neural rendering with the architecture principles behind modern language and video models. Unlike traditional neural rendering methods, Kaleido learns 3D purely in a data-driven way, without explicit 3D representations or structures. It acquires spatial understanding directly through large-scale video pretraining, then multi-view 3D data finetuning, inspired by how LLMs acquire textual common sense from large corpora before specialising in domains like coding. Through extensive ablations, we progressively modernised the architecture design and training strategies and tackled key scaling challenges in sequence-to-sequence generative rendering, arriving at a design that’s simple, versatile, and scalable. Kaleido significantly outperforms prior generative models in few-view settings, and remarkably is the first zero-shot generative method matches InstantNGP-level rendering quality in multi-view settings. We view Kaleido also as an alternative step towards world modeling that flexibly spans a spectrum of “realities": with many views, it faithfully reconstructs grounded reality; with fewer views, it imagines plausible unseen details. 🔗 Explore more results and paper:

Introducing Kaleido💮 from AI at Meta — a universal generative neural rendering engine for photorealistic, unified object and scene view synthesis. Kaleido is built on a simple but powerful design philosophy: 3D perception is a form of visual common sense. Following this idea, we formulate rendering purely as a sequence-to-sequence generation problem, successfully unifying neural rendering with the architecture principles behind modern language and video models. Unlike traditional neural rendering methods, Kaleido learns 3D purely in a data-driven way, without explicit 3D representations or structures. It acquires spatial understanding directly through large-scale video pretraining, then multi-view 3D data finetuning, inspired by how LLMs acquire textual common sense from large corpora before specialising in domains like coding. Through extensive ablations, we progressively modernised the architecture design and training strategies and tackled key scaling challenges in sequence-to-sequence generative rendering, arriving at a design that’s simple, versatile, and scalable. Kaleido significantly outperforms prior generative models in few-view settings, and remarkably is the first zero-shot generative method matches InstantNGP-level rendering quality in multi-view settings. We view Kaleido also as an alternative step towards world modeling that flexibly spans a spectrum of “realities": with many views, it faithfully reconstructs grounded reality; with fewer views, it imagines plausible unseen details. 🔗 Explore more results and paper:

Shikun Liu

22,389 views • 10 months ago

We just shipped v10 of Shopify POS. Major redesign. Faster. Smarter. And built for selling in person, your way. Here are the headlines: - Branded POS screens - New home screen layout & navigation - Redesigned cart - Smarter, faster search Fewer taps. Cleaner workflows. More control. If you’re not already selling in person, there’s never been a better time to start.

We just shipped v10 of Shopify POS. Major redesign. Faster. Smarter. And built for selling in person, your way. Here are the headlines: - Branded POS screens - New home screen layout & navigation - Redesigned cart - Smarter, faster search Fewer taps. Cleaner workflows. More control. If you’re not already selling in person, there’s never been a better time to start.

Harley Finkelstein

12,262 views • 1 year ago

Resilience is not built by one nation alone. To build a better future together, we must unite in our strength. Thanks to Council on Foreign Relations for inviting me to discuss how Canada and our allies can build a new rules-based order, and prosper in it.

Resilience is not built by one nation alone. To build a better future together, we must unite in our strength. Thanks to Council on Foreign Relations for inviting me to discuss how Canada and our allies can build a new rules-based order, and prosper in it.

Mark Carney

87,650 views • 10 months ago

We just open-sourced 3DGS render blender addon v4.0 ( It has been a while, but we are still updating this 3D Gaussian Splatting tool! In this version, we added a real-time rendering mode, so say goodbye to those laggy displays in the viewport! Here is an example: I imported a file that contains one million gaussians, super smooth.

We just open-sourced 3DGS render blender addon v4.0 ( It has been a while, but we are still updating this 3D Gaussian Splatting tool! In this version, we added a real-time rendering mode, so say goodbye to those laggy displays in the viewport! Here is an example: I imported a file that contains one million gaussians, super smooth.

Chris make some 3D scans

68,522 views • 10 months ago

the so101 + leslider frame i've been playing with the past few weeks wasn't just a random rig i built for fun it's what we at LiveKit built to benchmark and debug our infra (s/o livekit-portal), and today we're open-sourcing it so everyone can reproduce and share coolio results at home here is the github: i know this isn’t the first frame designed for evals, but it’s ours, and it’s open, and it’s with a LeSlider! along with this, we’ve released simulations files (URDF, MJCF, and USD), as well as a sample RL env in MJLab for pick-and-place tasks we want people to have fun, being close to the frontier as much as possible, and we want people to know it shouldn’t take thousands of dollars in equipment to reproduce some of the most amazing results the academic world has to offer personally, I’m going to deploy sim2real and real2sim2real pipelines on this soon, for fun, so look out for that.

the so101 + leslider frame i've been playing with the past few weeks wasn't just a random rig i built for fun it's what we at LiveKit built to benchmark and debug our infra (s/o livekit-portal), and today we're open-sourcing it so everyone can reproduce and share coolio results at home here is the github: i know this isn’t the first frame designed for evals, but it’s ours, and it’s open, and it’s with a LeSlider! along with this, we’ve released simulations files (URDF, MJCF, and USD), as well as a sample RL env in MJLab for pick-and-place tasks we want people to have fun, being close to the frontier as much as possible, and we want people to know it shouldn’t take thousands of dollars in equipment to reproduce some of the most amazing results the academic world has to offer personally, I’m going to deploy sim2real and real2sim2real pipelines on this soon, for fun, so look out for that.

Binh Pham

18,158 views • 26 days ago

Finally getting to share one of my favorite projects. ICLR Oral! 🏆 It’s so strange how rigid video tokenization is. Think about it: why should a still landscape cost the same amount of tokens as a busy street? We built InfoTok. We went back to basics with Shannon’s information theory to make tokens "adaptive" in a principled way. Its 2.3x better compression and 11x faster inference demonstrates the magic of the old-school theory ✨ Check it out:

Finally getting to share one of my favorite projects. ICLR Oral! 🏆 It’s so strange how rigid video tokenization is. Think about it: why should a still landscape cost the same amount of tokens as a busy street? We built InfoTok. We went back to basics with Shannon’s information theory to make tokens "adaptive" in a principled way. Its 2.3x better compression and 11x faster inference demonstrates the magic of the old-school theory ✨ Check it out:

Haotian Ye

49,626 views • 4 months ago

Is Google taking initial steps to enhance Street View? For some reason, Street View seems stuck in technology that feels outdated. I wonder if we'll see such improvements on the product side. Also, note how much better it performs in all aspects compared to Zip-NeRF in their presented material. It offers more details and fewer artifacts. Great work! "LODGE: Level-of-Detail Large-Scale Gaussian Splatting with Efficient Rendering" Contributions: • We propose a novel LOD representation for 3DGS which, unlike previous methods [27, 28, 17], does not recompute the list of used Gaussians at each frame. This allows for acceleration and compaction, enabling the rendering of large-scale scenes even on mobile devices. • We design a strategy to automatically select optimal hyperparameters for splitting LODs, whereas most other methods require manual tuning of hyperparameters for each 3D scene. • To further accelerate rendering, we split the scene into chunks and pre-compute sets of active Gaussians per chunk. • Finally, we introduce a novel opacity interpolation scheme to produce visually pleasing rendering and eliminate artifacts when transitioning between chunks.

Is Google taking initial steps to enhance Street View? For some reason, Street View seems stuck in technology that feels outdated. I wonder if we'll see such improvements on the product side. Also, note how much better it performs in all aspects compared to Zip-NeRF in their presented material. It offers more details and fewer artifacts. Great work! "LODGE: Level-of-Detail Large-Scale Gaussian Splatting with Efficient Rendering" Contributions: • We propose a novel LOD representation for 3DGS which, unlike previous methods [27, 28, 17], does not recompute the list of used Gaussians at each frame. This allows for acceleration and compaction, enabling the rendering of large-scale scenes even on mobile devices. • We design a strategy to automatically select optimal hyperparameters for splitting LODs, whereas most other methods require manual tuning of hyperparameters for each 3D scene. • To further accelerate rendering, we split the scene into chunks and pre-compute sets of active Gaussians per chunk. • Finally, we introduce a novel opacity interpolation scheme to produce visually pleasing rendering and eliminate artifacts when transitioning between chunks.

MrNeRF

62,564 views • 1 year ago

The craft of engineering is rapidly changing. At Ramp, we built our own background coding agent to accelerate faster. We call it Inspect. It wrote 30% of merged frontend + backend PRs in the past week. It’s powered by OpenCode, Modal and Cloudflare Developers. It runs fully in the cloud, and starts in seconds, letting every builder work at the speed of thought, no setup required. Today, we’re open sourcing the full blueprint so anybody can build their own Inspect. Just give our spec to your current coding agent, and let it build your new favourite.

The craft of engineering is rapidly changing. At Ramp, we built our own background coding agent to accelerate faster. We call it Inspect. It wrote 30% of merged frontend + backend PRs in the past week. It’s powered by OpenCode, Modal and Cloudflare Developers. It runs fully in the cloud, and starts in seconds, letting every builder work at the speed of thought, no setup required. Today, we’re open sourcing the full blueprint so anybody can build their own Inspect. Just give our spec to your current coding agent, and let it build your new favourite.

Zach Bruggeman

802,543 views • 6 months ago

💜 bbuu was actually a song that had been sleeping in our work folder, we're waiting for the right moment to release it, as soon as the moment we heard about our collab with sanrio, we instantly thought, "this is the song!" so we refined it further and that's how bbuu came to be

💜 bbuu was actually a song that had been sleeping in our work folder, we're waiting for the right moment to release it, as soon as the moment we heard about our collab with sanrio, we instantly thought, "this is the song!" so we refined it further and that's how bbuu came to be

jess

19,399 views • 7 months ago

Every AI agent scraping the web right now is running full chrome to do it. rendering a UI nobody looks at, for a browser that never displays a single pixel. lightpanda skips that entirely. written from scratch in zig, not a chromium fork, and it's 11x faster with 9x less memory than chrome in headless mode. → one-line docker install → drops into playwright, puppeteer, and chromedp through cdp, port 9222 11.8k stars for something that basically asks why you were running a full desktop browser to fetch json.

Every AI agent scraping the web right now is running full chrome to do it. rendering a UI nobody looks at, for a browser that never displays a single pixel. lightpanda skips that entirely. written from scratch in zig, not a chromium fork, and it's 11x faster with 9x less memory than chrome in headless mode. → one-line docker install → drops into playwright, puppeteer, and chromedp through cdp, port 9222 11.8k stars for something that basically asks why you were running a full desktop browser to fetch json.

Oliver Prompts

38,715 views • 12 days ago

A Midjourney engineer just open-sourced a way to lay out entire web pages without CSS. Not a framework. Not a library that wraps the DOM. A pure TypeScript text measurement algorithm that bypasses browser reflow entirely. 600x faster. Why it matters: AI-generated interfaces need to lay out text dynamically. The browser wasn't built for that. CSS reflow was designed for humans writing static pages, not agents generating UIs on the fly. This exists because agents need rendering that doesn't depend on a 30-year-old pipeline. The AI-era stack is being rebuilt from scratch. One library at a time. P.S. Such a fun to play with that!

A Midjourney engineer just open-sourced a way to lay out entire web pages without CSS. Not a framework. Not a library that wraps the DOM. A pure TypeScript text measurement algorithm that bypasses browser reflow entirely. 600x faster. Why it matters: AI-generated interfaces need to lay out text dynamically. The browser wasn't built for that. CSS reflow was designed for humans writing static pages, not agents generating UIs on the fly. This exists because agents need rendering that doesn't depend on a 30-year-old pipeline. The AI-era stack is being rebuilt from scratch. One library at a time. P.S. Such a fun to play with that!

Paweł Huryn

238,863 views • 4 months ago

We took a 30B model and split it in two to write tokens in parallel instead of one at a time. Introducing Nemotron-Labs-TwoTower: a diffusion language model from NVIDIA Research adapted from Nemotron-3-Nano-30B-A3B. Here’s how it works: one half holds the context, the other writes the tokens, with both reusing the pretrained model instead of training a new one from scratch. We found it kept 98.7% of the original model’s quality at 2.42× faster generation.

We took a 30B model and split it in two to write tokens in parallel instead of one at a time. Introducing Nemotron-Labs-TwoTower: a diffusion language model from NVIDIA Research adapted from Nemotron-3-Nano-30B-A3B. Here’s how it works: one half holds the context, the other writes the tokens, with both reusing the pretrained model instead of training a new one from scratch. We found it kept 98.7% of the original model’s quality at 2.42× faster generation.

NVIDIA AI

759,905 views • 1 month ago

😍 Widgets concept in HOT Wallet You've been asking for it for a long time, and we've wanted to do it for a long time. So far, this is just a concept, but soon we will add this functionality to the wallet to make your experience even better. In addition to the widgets developed by us, we also plan to support widgets from third-party developers. If you are a project or an influencer, this can be a great opportunity to add value to your audience. Contact us to find out the details.

😍 Widgets concept in HOT Wallet You've been asking for it for a long time, and we've wanted to do it for a long time. So far, this is just a concept, but soon we will add this functionality to the wallet to make your experience even better. In addition to the widgets developed by us, we also plan to support widgets from third-party developers. If you are a project or an influencer, this can be a great opportunity to add value to your audience. Contact us to find out the details.

Nik

23,727 views • 1 year ago

Today we're launching Sailboxes: the first cloud environment purpose-built for long-horizon AI agents. Sailboxes are full machines with persistent state that auto-sleep, built for agents that live forever, priced so you only pay for what they actually use. From $0.015 per active vCPU-hour, over 70% cheaper than other providers. Under the hood: novel architecture from Nirvik Baruah and Charley Cunningham that live-migrates VMs based on actual resource usage. To show how robust it is, we ran a Minecraft server in a Sailbox and forced it to migrate every 2 minutes--players never noticed.

Today we're launching Sailboxes: the first cloud environment purpose-built for long-horizon AI agents. Sailboxes are full machines with persistent state that auto-sleep, built for agents that live forever, priced so you only pay for what they actually use. From $0.015 per active vCPU-hour, over 70% cheaper than other providers. Under the hood: novel architecture from Nirvik Baruah and Charley Cunningham that live-migrates VMs based on actual resource usage. To show how robust it is, we ran a Minecraft server in a Sailbox and forced it to migrate every 2 minutes--players never noticed.

Sail Research

56,545 views • 23 days ago

Reya is bringing high performance trading to the Ethereum L1 in a way not seen since DeFi Summer: atomically composable with the L1 ecosystem. We used to call it money legos. That means more ways for users to combine DeFi applications. More ways to hedge, express views and find edges as the system evolves. This is a new type of trading layer. One built for a new world. One built for you.

Reya is bringing high performance trading to the Ethereum L1 in a way not seen since DeFi Summer: atomically composable with the L1 ecosystem. We used to call it money legos. That means more ways for users to combine DeFi applications. More ways to hedge, express views and find edges as the system evolves. This is a new type of trading layer. One built for a new world. One built for you.

Reya

29,499 views • 7 months ago

I'm proud to share that Glean has surpassed $300M ARR, just five months after crossing $200M and growing ~3x over the past 15 months. This is an exciting milestone for Glean, and it's a signal about where the enterprise AI market is heading. We’ve long believed the real challenge in enterprise AI is not access to models. It is grounding AI in how a company actually works: its people, knowledge, workflows, permissions, and systems. That’s even clearer now. The companies creating real value with AI are not just adopting better models. They are building systems that understand their business well enough to deliver reliable outcomes at scale. That is the real moat, and it is what we’ve been building at Glean: an unrivaled context layer for enterprise AI. That context has to work across the business, not just inside a single team or use case. We see that in how customers adopt Glean: more than 85% use it across five or more job functions. It also has to meet the security and governance demands of complex enterprises. We see that in who is choosing Glean: our Fortune 500 customer count nearly doubled year over year. And it has to make economic sense as usage grows. In our recent benchmark with Claude Cowork, Glean was preferred roughly 2.5x as often as off-the-shelf MCP tools and used 30% fewer tokens on average. Better context improves both quality and efficiency. I enjoyed talking with CNBC's Deirdre Bosa about this broader shift. In enterprise AI, the winners will not be defined by better models alone. They will be defined by who builds the strongest foundation for enterprise context. Thank you to our customers, partners, and team for helping us build the future of enterprise AI.

I'm proud to share that Glean has surpassed $300M ARR, just five months after crossing $200M and growing ~3x over the past 15 months. This is an exciting milestone for Glean, and it's a signal about where the enterprise AI market is heading. We’ve long believed the real challenge in enterprise AI is not access to models. It is grounding AI in how a company actually works: its people, knowledge, workflows, permissions, and systems. That’s even clearer now. The companies creating real value with AI are not just adopting better models. They are building systems that understand their business well enough to deliver reliable outcomes at scale. That is the real moat, and it is what we’ve been building at Glean: an unrivaled context layer for enterprise AI. That context has to work across the business, not just inside a single team or use case. We see that in how customers adopt Glean: more than 85% use it across five or more job functions. It also has to meet the security and governance demands of complex enterprises. We see that in who is choosing Glean: our Fortune 500 customer count nearly doubled year over year. And it has to make economic sense as usage grows. In our recent benchmark with Claude Cowork, Glean was preferred roughly 2.5x as often as off-the-shelf MCP tools and used 30% fewer tokens on average. Better context improves both quality and efficiency. I enjoyed talking with CNBC's Deirdre Bosa about this broader shift. In enterprise AI, the winners will not be defined by better models alone. They will be defined by who builds the strongest foundation for enterprise context. Thank you to our customers, partners, and team for helping us build the future of enterprise AI.

Arvind Jain

280,572 views • 2 months ago

Hey geoff! We saw that you publicly sold some tokens and it caused a huge price dip~!! Taking profits is totally fine (no shame there), but maybe there's a better way to sell your tokens with more privacy..? We're building a solution for this, called Ghost Swap. Ghost Swap is built on top of a privacy mixer pool and brings new utility to privacy-preserving capital. You can deposit SOL / any token into Ghost Swap and use your shielded tokens to interact with public liquidity, such as swapping. If you’re open to it, we’d love to get you in as an early tester to try it out when it goes live. 🧡

Hey geoff! We saw that you publicly sold some tokens and it caused a huge price dip~!! Taking profits is totally fine (no shame there), but maybe there's a better way to sell your tokens with more privacy..? We're building a solution for this, called Ghost Swap. Ghost Swap is built on top of a privacy mixer pool and brings new utility to privacy-preserving capital. You can deposit SOL / any token into Ghost Swap and use your shielded tokens to interact with public liquidity, such as swapping. If you’re open to it, we’d love to get you in as an early tester to try it out when it goes live. 🧡

Monero-Chan

51,685 views • 6 months ago

Today we’re open-sourcing Stable Audio Open Small, a 341M-parameter text-to-audio model optimized to run entirely on Arm CPUs. This means 99% of smartphones can now generate music-production samples in seconds, right on-device with no internet required. Built for fast, on-the-go creation, it turns your next quick idea into up to 11 seconds of audio. Generate drum loops, foley, riffs, and textures right where you are. No cords 🔌 just chords 🎹 You can learn more here:

Today we’re open-sourcing Stable Audio Open Small, a 341M-parameter text-to-audio model optimized to run entirely on Arm CPUs. This means 99% of smartphones can now generate music-production samples in seconds, right on-device with no internet required. Built for fast, on-the-go creation, it turns your next quick idea into up to 11 seconds of audio. Generate drum loops, foley, riffs, and textures right where you are. No cords 🔌 just chords 🎹 You can learn more here:

Stability AI

94,796 views • 1 year ago