First fully ML-framework-free 3D Gaussian Splatting implementation in LichtFeld... Studio. I’ve completed the migration of the full training pipeline to a custom CUDA-based tensor library. No PyTorch, no LibTorch, no autograd. Every gradient is implemented by hand, either through CUDA kernels or minimal abstractions on top. This makes it the first full training setup for 3D Gaussian Splatting with zero dependencies on existing ML frameworks. It’s not just about independence, it's about control! We now manage every byte of GPU memory, which opens the door to tighter optimization and finer performance tuning. The framework footprint is minimal, without pulling in gigabytes of ML runtime code that was never designed for real-time or graphics-driven applications. A few modules, such as the metrics and 3DGUT interfaces, are still being ported, and some operations are temporarily naïve, so performance is not yet on par with master. But this refactor lays the groundwork for: - A fully self-contained binary - Fine-grained memory optimization - Easier experimentation without the weight of an ML stack We’re getting close.show more

Apple just trained a 3D Gaussian head reconstruction model... on 10,000+ subjects. Feed-forward. No test-time optimization. New identity in, reconstructed Gaussian head out. The UV-parameterized Gaussian representation decouples the number of Gaussians from the number and resolution of input images, making it practical to train with many high resolution views. And the heads are not just static either: text-conditioned identity generation, plus blendshape-driven latent animation across identities. We've been building in the 3D Gaussian Splatting space for a while. The gap between "research demo" and "works on real people at scale" is closing fast.show more

Two weeks ago I fixed one of my teeth... with algorithms I wrote a couple of years ago! I got hooked by 3D scanning when I started to work for a software shop in Zurich that was programming 3D computational geometry algorithms for denture scanning to produce crowns (and more). Back then, a typical reconstruction pipeline was like: scan the patient’s teeth using an intraoral scanner, reconstruct the surface mesh, design the restoration digitally, and finally mill the crown out of ceramic. We were working mostly with point clouds and meshes, but it wasn’t just math, it was craftsmanship translated into a digital process. Every micron mattered. You could literally see how a good algorithm meant a better fit in someone’s mouth. Gaussian Splatting isn’t about surface reconstruction, it’s about appearance reconstruction. It doesn’t care about explicit topology, it captures how light interacts with the scene. In a sense, it’s the opposite philosophy of the dental world: instead of modeling what the object is, it models how the object looks. 3D Gaussian Splatting enables applications like training self driving cars, teaching robots to understand their environment, creating virtual worlds, or monitoring real sites. It represents scenes as millions of small Gaussians rendered in real time without the need for meshes or textures. Coming from a world where precision geometry was everything, this shift felt natural. It’s still about reconstruction, but with a different goal: not manufacturing a perfect object, but reproducing how the world actually looks. Two weeks ago I got my first dental crown, made with the same software, reconstruction algorithms, and Swiss precision I once helped develop. I haven’t worked there in two years, but sitting in that chair and seeing the process from the other side was a proud moment. It reminded me why I love this field.show more

🚨 Anthropic committed up to 1M TPU chips for... Claude. Openai is leasing TPUs for chatgpt inference. Here's How kernels work on TPUs (deep dive 2/6 by emi) pallas is Google's answer to kernel writing. a python kernel SDK built on JAX. still very experimental (jax.experimental.pallas). on TPU it compiles through mosaic; on GPU it lowers to triton. if you know CUDA, the syntax will feel familiar but the execution model is completely different. in CUDA, grid=(4,4) launches 16 blocks running simultaneously across SMs. in pallas, those 16 iterations run one after another in lexicographic order. no threads. no warps. no blocks. no occupancy tuning. a TPU is a sequential machine with a very wide vector register — more like a CPU than a GPU. performance comes from width: a 128x128 systolic array doing matmul and an 8x128 SIMD vector unit doing everything else. maximum parallelism on chip: 2, one per TensorCore in megacore mode. three concepts replace CUDA's thread/block/grid hierarchy. Refs are mutable memory references. because execution is sequential, each iteration safely accumulates without atomics. in CUDA you'd need atomics or a separate reduction pass. the memory model is also very different from NVIDIA's. zero hardware caches. VMEM is 32-128 MiB of software-managed scratchpad — 500-1000x larger than GPU shared memory per SM. all data must be explicitly DMA'd from HBM to VMEM before any computation touches it. four levels: HBM → VMEM → VREGs → MXU/VPU, plus SMEM for scalar control data. every byte of data movement is your responsibility. this is like CUDA shared memory except it's 500x bigger and there's no cache fallback. pipelining is mandatory. without double-buffering HBM→VMEM transfers, the MXU just stalls waiting for data. this is the single most important optimization on TPU. and because grid execution is sequential and deterministic, consecutive iterations that need the same input block skip the redundant HBM transfer automatically, impossible on GPU where block execution order is undefined. the compilation pipeline is unlike anything in this series: python → jaxpr → stableHLO → XLA HLO (71+ optimization passes) → LLO (78+ passes) → 322-bit VLIW bundles. the compiler packs instructions for scalar, vector, matrix, and DMA units into a single 322-bit word. everything in that bundle executes in parallel, with no runtime scheduling.show more

Despite the Framer hype, we still get tons of... requests for custom Next.js + Tailwind sites. Because most founders and their teams don't want to learn another tool. They want to own their code, not rent it. Just shipped a landing page in Next.js for a SaaS. Their reasoning: - Full control over performance optimization - No vendor lock-in - Developers can maintain it without learning Framer - Custom functionality without workarounds - Better SEO control Sure, Framer is faster to build. But he no-code movement sold everyone on "anyone can build websites now" and forgot that someone still needs to maintain them. What's your take: Custom code for control or no-code for speed?show more

The West is not dying. It is being killed,... and the names of the traitors are known. They occupy our capitals, infest our courts, pollute our newsrooms, and preach in our churches. They open the gates, kneel before the foreigner, and smirk as their own blood is driven from the land. They mock the fallen, defile the heroic, and spit on the blood that raised every city worth defending. They are not misguided. They are not mistaken. They are the enemy. They must be treated as such. For too long, we have been ruled by cowards, “men without chests,” by merchants loyal to nothing but the dollar, by liars who speak of progress while presiding over decay. A new generation now rises, armed not with apologies but with the fire of remembrance, with the memory of what we once were and the will to become greater still. We do not ask permission. We do not seek approval. We will reclaim what is ours, because no one else will. Victory will not come through debate. It will come through discipline, through will, through the unbreakable decision to endure, to outlast, and to return to the excellence and greatness that befit our people. We do not need millions. We require only a vanguard: men of loyalty, endurance, and resolve, hardened by truth and unmoved by fear. I say this not for approval, nor is it offered in hope of a reply, but in the spirit of doing what must be done. It is a promise made in full knowledge of what must come. The time of submission draws to a close. The age of reconquest begins. Let the traitors tremble. Let the weak, the feckless, and the unworthy fall away. The future belongs to those with the strength and the daring to seize it.show more

3D Gaussian Splatting for Real-Time Radiance Field Rendering paper... page: Radiance Field methods have recently revolutionized novel-view synthesis of scenes captured with multiple photos or videos. However, achieving high visual quality still requires neural networks that are costly to train and render, while recent faster methods inevitably trade off speed for quality. For unbounded and complete scenes (rather than isolated objects) and 1080p resolution rendering, no current method can achieve real-time display rates. We introduce three key elements that allow us to achieve state-of-the-art visual quality while maintaining competitive training times and importantly allow high-quality real-time (>= 30 fps) novel-view synthesis at 1080p resolution. First, starting from sparse points produced during camera calibration, we represent the scene with 3D Gaussians that preserve desirable properties of continuous volumetric radiance fields for scene optimization while avoiding unnecessary computation in empty space; Second, we perform interleaved optimization/density control of the 3D Gaussians, notably optimizing anisotropic covariance to achieve an accurate representation of the scene; Third, we develop a fast visibility-aware rendering algorithm that supports anisotropic splatting and both accelerates training and allows realtime rendering. We demonstrate state-of-the-art visual quality and real-time rendering on several established datasets.show more

Introducing ml-intern, the agent that just automated the post-training... team Hugging Face It's an open-source implementation of the real research loop that our ML researchers do every day. You give it a prompt, it researches papers, goes through citations, implements ideas in GPU sandboxes, iterates and builds deeply research-backed models for any use case. All built on the Hugging Face ecosystem. It can pull off crazy things: We made it train the best model for scientific reasoning. It went through citations from the official benchmark paper. Found OpenScience and NemoTron-CrossThink, added 7 difficulty-filtered dataset variants from ARC/SciQ/MMLU, and ran 12 SFT runs on Qwen3-1.7B. This pushed the score 10% → 32% on GPQA in under 10h. Claude Code's best: 22.99%. In healthcare settings it inspected available datasets, concluded they were too low quality, and wrote a script to generate 1100 synthetic data points from scratch for emergencies, hedging, multilingual etc. Then upsampled 50x for training. Beat Codex on HealthBench by 60%. For competitive mathematics, it wrote a full GRPO script, launched training with A100 GPUs on watched rewards claim and then collapse, and ran ablations until it succeeded. All fully backed by papers, autonomously. How it works? ml-intern makes full use of the HF ecosystem: - finds papers on arxiv and reads them fully, walks citation graphs, pulls datasets referenced in methodology sections and on - browses the Hub, reads recent docs, inspects datasets and reformats them before training so it doesn't waste GPU hours on bad data - launches training jobs on HF Jobs if no local GPUs are available, monitors runs, reads its own eval outputs, diagnoses failures, retrains ml-intern deeply embodies how researchers work and think. It knows how data should look like and what good models feel like. Releasing it today as a CLI and a web app you can use from your phone/desktop. CLI: Web + mobile: And the best part? We also provisioned 1k$ GPU resources and Anthropic credits for the quickest among you to use.show more

HE BUILT A $10,000-TIER ANIMATED SITE WITH CLAUDE CODE... - FOR THE COST OF A SUBSCRIPTION What's on screen isn't a landing page with a parallax background It's a fully interactive, scroll-driven site with real-time 3D rendered in the browser What's actually on the page: > 3D product models rotating and reacting to scroll in real time via WebGL > Smooth hover interactions and transitions - no hand-coded keyframes > Cinematic minimal aesthetic that got featured on Awwwards > Typography layering, editorial layouts, everything assembled in one session What it normally takes: > A 3D artist, a motion designer, and a frontend developer > Weeks of handoffs - modelling, exporting, wiring animations, layout, copy > Six separate systems integrated by hand That pipeline was the moat. It's what justified the invoice The price gap: > Studio build at this level: $5,000-10,000+ > Your cost: a Claude subscription Timeline: weeks of production -> a single session Full walkthrough in the article belowshow more

This is not a Chinese family going on vacation... with their dog. This is a dog likely bought from a wet market for meat. They have tied the poor animal to the back of their car and are driving on the highway as if he is just luggage. He has no idea what is waiting for him. That is what makes this so heartbreaking. This dog is still alive, still trusting, still innocent — and yet he is being treated with no care, no mercy, and no dignity. Dog meat eating is still a cruel reality in parts of China. Many of these dogs are stolen pets or helpless strays. No dog deserves this fate. Dogs are companions. They are not food.show more

HTML Artifacts are a big part of how I... work with agents now. Artifacts can be more than just static files. When combined with agents, they can take action or help you take action. This unlocks all kinds of interesting ways to work with agents. This is clearly the future. Check out this writing and scheduler artifact I built in a few minutes. It uses a bit of HTML and JS. All the data is in markdown (Obsidian vaults), so the agent can access and modify it at any time. No DB needed. No sophisticated functionalities. The agent decides all that for me based on the skills, context, and memory it has access to. The best part about this simple stack is that all the important information stays with me. This has allowed me to build a recursive self-improving system and automations that can better tap into coding agents like Codex or Claude Code. I could have paid or built an entire app for scheduling posts, and there are so many of them out there. But I don't need to. I've realized a simple artifact does the job. And the simplicity of it is actually an advantage. Very little maintenance for very high returns on personalization, time, and efficiency. The other benefit of this is that I can add features as I please. That level of personalization feels magical, and we should all be pursuing more of it. All of this just keeps compounding. Of course, this example is just about writing. But I have similar artifacts for research, design, experimentation, evaluation, and so much more. And no, I didn't actually publish the post example I shared in the clip. It was just for demonstration purposes. I actually spend more time than this when writing together with agents. Lastly, having built my own agent orchestrator tool has made me realize that simplifying the tool stack is a superpower. If you are curious about how all this works, I will do a live session next week:show more

Disappointed with your ICLR paper being rejected? Ten years... ago today, Sergey and I finished training some of the first end-to-end neutral nets for robot control 🤖 We submitted the paper to RSS on January 23, 2015. It was rejected for being "incremental" and "unlikely to have much impact" Our resubmission to NeurIPS was also rejected It now has >4,000 citations (and more importantly, end-to-end training is widely accepted!) It's also cool to think about what's changed and what's the same -- - The network was 92k parameters and trained on ~15 minutes of data - The code was a combination of matlab, caffe, ROS, a custom CUDA kernel for speed, and a low-level 20 Hz controller in C++, all talking to each other. ROS+matlab was as bad as it sounds. - We pre-trained the encoder and did inference off-board on a workstation with a larger GPU. - We were paranoid about varying lighting messing up the network, so we did all the experiments after sunset (so long nights running experiments on the robot past 3 am) Now, we have manipulation policies that are far more dextrous, far more generalizable, and maybe on the cusp of breaking into the real world. :) (the paper:show more

Stablecoin regulation has arrived, and here’s why it matters... for merchants: After years of uncertainty, stablecoins are now supported by clear regulations in the US, EU, and UK. They are no longer seen as just a speculative technology. They are now a practical, regulated option for payments. For merchants, this opens the door to faster settlements, lower costs, improved cash flow, and a smoother checkout experience. With regulatory clarity in place, businesses can adopt stablecoins without hesitation. The opportunity is real. The path is clear. Now, it’s all about execution. Read the full blog:show more

Exciting Milestone: Our First CEX Listing with MEXC! We’re... thrilled to announce that VentureMind AI ($VNTR) is officially listed on mexc_listings ! Choosing our first centralized exchange partner was a big decision, and we’re excited to team up with MEXC to bring $VNTR to a wider audience. This is a historic moment for us as the first AI incubator project launched on the Seedify platform to achieve a CEX listing. We couldn’t have reached this point without Seedify’s guidance, helping us navigate the journey with confidence and clarity. This partnership is just the beginning! It gives easier access to $VNTR and introduces the first onramp to purchase our token using fiat stablecoins, making it more accessible to a global audience. And we’re just getting started! Additional exchange listings are already in the works, and we can’t wait to share more milestones as we expand the reach and utility of $VNTR. To our incredible community, thank you for your unwavering support. This is only the start of an exciting journey, and we’re so grateful to have you with us every step of the way!show more

In the summer of 2023, I cold emailed Jensen... Huang and asked to capture a NeRF of him at SIGGRAPH. He responded in about an hour and said yes. A radiance field is, in the simplest terms, akin to a 3D photograph. A moment in time, so completely reconstructed that you can move through it and see it from angles the original cameras never occupied. NeRFs were the original method. Gaussian splatting, which debuted at that same SIGGRAPH, has since become the dominant form of radiance field. I called my late friend James, who told me we needed to begin practicing immediately. We ran capture after capture for weeks until we consistently got the capture time down to ~30 seconds with one camera. Later, in a hallway at the LA Convention Center during SIGGRAPH, I captured the portrait you're seeing now, a full 360° gaussian splat of Jensen, rendered here as a 2D flythrough. Afterward, I continued the conversation with him and members of his team to make the case for radiance fields as a foundational representation for imaging. To my surprise, they listened. Three years later, NVIDIA has several works, including NuRec, fVDB, 3DGRUT, and gsplat all utilizing radiance fields. The landscape has evolved enough that the reasoning is obvious. Gaussian splatting has begun to ship across some of the world’s largest industries, including autonomous vehicles, AEC, geospatial, media and entertainment, robotics, e-commerce, hospitality. It’s become clear that lifelike 3D is here to stay. And yet I think we will look back and be disappointed by how late we started taking 3D portraits of the people around us, just like how we have sparse 2D photos of our grandparents and great grandparents. We have billions of photographs of the people we know and love, but almost no radiance fields of them. I'll be returning to SIGGRAPH in LA where this was initially captured three years ago, with the landscape looking significantly different. Radiance fields are more under deployed than ever relative to what they can do. I'm excited for the future of imaging, and for 2D to transition into 3D. I have a few things up my sleeve that I think will make that case plainly.show more

🇺🇸 DALLAS IS ABOUT TO BECOME THE CAPITAL OF... THE AMERICA FIRST MOVEMENT! For the first time in party history, Republicans are holding a midterm convention. Not a debate stage. Not a press conference. A full blown convention, September 9th and 10th, right in the heart of Dallas. The Democrats aren't ready for this! Trump called it exactly what it is. "It has never been done before, and will be a truly Historic Event." Think about that for a second. Presidential conventions happen every four years like clockwork. Midterm conventions do not happen at all, because most parties do not have anything worth celebrating in the off year. The GOP just decided they do. RNC Chairman Joe Gruters is already calling it Trumpapalooza. That is not a typo. That is the energy level we are talking about. This is not just a rally. It is a two day showcase of the Great American Comeback. No tax on tips. No tax on overtime. Falling oil prices while the administration denuclearizes Iran. A border that finally has a lock on the door. Trump is not asking Republicans to imagine the wins. He is putting them on a stage in Texas and pointing at them. And Texas is not a random choice. It is the epicenter of this year's fight for Congress, with Ken Paxton battling for Senate and multiple House seats hanging in the balance. Dallas Mayor Eric Johnson called it exactly right, an event that will "energize our party, strengthen the conservative movement, and help build momentum." Now look across the aisle. Democrats floated the idea of their own midterm gathering. Then they quietly shelved it. No unifying message. No standout headliner. No comeback story to tell voters, because they do not have one. That is the difference in one sentence. Republicans are throwing a party because they have something to celebrate. Democrats are staying home because they do not. Midterms usually punish the party in power. Trump just decided to rewrite that rule in Texas, in front of the cameras, with the whole country watching. Buckle up. Trumpapalooza is coming, and the other side has nothing on the calendar to answer it with.show more

Alaska. Putin and Trump. The meeting that was both... awaited and feared. The outcome. By Roman Alekhin To discuss this further, the key point to understand is this: Two and a half hours behind closed doors is too little for leaders of nuclear powers if they are just starting a conversation—but too much if they were merely posing for cameras. This means the real work happened earlier—quiet, working discussions “behind the door” that never made it into the official communiqué. These are what stopped Trump from imposing secondary sanctions; this is where the exchange of conditions took place, terms that won’t be disclosed until both leaders decide the time is right. I suspect Syria was discussed, as well as the “Trump Bridge” (instead of Zangezur), the Middle East, and much more that isn’t yet visible to the naked eye. The public part? Pure theater. The point wasn’t to negotiate in front of the world or sign something, like Trump did with Armenia and Azerbaijan—minor players, important only tactically. The point was to send a signal: the presidents of Russia and the U.S. are shaking hands and smiling again, no knives behind their backs. This signals that a new reality has arrived, one that Europe, Ukraine, and everyone accustomed to building their worldview around the idea of “Russia’s isolation” will now have to reckon with—including those within our own countries. Trump’s comments on Fox News were deliberately vague—he’s a master at leaving room for maneuver. But the key takeaway is clear: the pressure will now shift from Moscow to Kyiv and Brussels. This is evident even in the final format—not a word about a “no-conditions” ceasefire, which was Europe and Zelensky’s main demand. This means the discussion on a final peace has been postponed, but within a clear framework: Zelensky must exit the war in a way that lets Europe save face. Russia’s red lines have long been clear: non-aligned status for Ukraine, return of Kherson, Zaporizhzhia, Donetsk, and Luhansk regions within their administrative borders (not the frontlines), lifting of sanctions, and real democracy in Ukraine without persecution of the Russian language or the Orthodox Church. These conditions will prolong the conflict for some time. Zelensky will need to stage a fighting retreat to the administrative borders—this will allow Europe to save face. But the logic is already visible: Ukraine’s defeat will be framed not as a “crushing” but as a “peaceful settlement.” Six months—that’s the timeframe in which we’ll see the dynamics. If the front accelerates, the deal is working. If it stalls, not all pieces are in place yet. But the main thing is already done: Russia is back in the game, not through gray negotiating formats but through a handshake on American soil—even if it’s chilly Alaska. (Though we have warm ties to the region, as seen in the wreath-laying for the “Heroes of ALSIB” and the meeting with Archbishop Alexy of Sitka and Alaska.) For the world, this is an image where smiles are worth more than signatures. For Trump—a chance to show he’s the only one who can “make peace.” For Putin—a symbolic victory: Russia is not a besieged enemy but an equal player. For Europe and Ukraine—the beginning of a painful new phase where they’ll have to accept the inevitable. In these talks, there were no winners or losers in Alaska. In Alaska, both presidents won, while those who weren’t there—lost, or at the very least, didn’t win.show more

Say hello to Boojum 👋: zkSync Era’s new high-performance... proof system for radical decentralization. Boojum is an upgrade that will transition zkSync Era to a STARK-powered proof system, providing world-class performance on consumer-grade hardware. 💡 Learn more: TL;DR 👇 Boojum is the name of our Rust-based cryptographic library, which we use to implement the upgraded version of the ZK circuits for zkSync Era and the ZK Stack. The name Boojum was inspired by Lewis Carroll's poem "The Hunting of the Snark," where the Boojum represents the most fearsome kind of Snark. We intentionally designed zkSync Era in a way that cryptographic upgrades can be made without a regenesis, meaning that the Boojum upgrade won’t cause any user disruptions. Why Boojum❓ From day one, zkSync’s mission is to advance personal freedom for all — making digital self-ownership universally accessible by building a blockchain network that is trustless, secure, permissionless, affordable, easy to use, resilient and limitlessly scalable. Boojum plays an important role in advancing this mission by delivering: 1. World-class performance zkSync Era’s current SNARK-based proof system is effective today, but it won’t scale to the volume that we envision for hyperchains. zkSync Era’s sequencer can already process over 100 TPS; Boojum orders of magnitude improvements to performance complements this well. 2. Reduced hardware requirements for decentralization Our long-term goal is to enable user-powered, decentralized proof generation. Boojum represents a breakthrough in this direction — with the prover running on consumer-grade GPUs requiring only 16 GB GPU RAM. Boojum’s Journey to Mainnet 🚴🏽‍♀️ Boojum is now live on Mainnet, generating and verifying ‘shadow proofs’ today with real production data so that we can carefully test the system ahead of fully migrating. Today, we’re also open-sourcing the repo; if you’d like to take a look, you can find it here 👇 This is the first of a series of posts on Boojum. We will provide updates on our progress, including more details on implementation, security, and performance. Watch here for more, anon ∎show more