I'll always root for a team that open-sources its... best work, and Robbyant just did it properly. Robbyant, Ant Group's embodied-AI company, released LingBot-Vision, a vision foundation model for robots, and the part I love is the data. They trained it on 161M images, filtered down from 2B raw ones and mostly pulled straight from the open web, with no human labels, no edge detectors, no depth sensors anywhere in the loop. It learns the exact edges of objects from raw pixels. That's roughly a tenth of the data DINOv3 saw, and under a third of the training. And it shows in the results. On depth, working out how far away things are, the 1B model edges out a 7B on NYU-Depth. It also powers LingBot-Depth 2.0, which reads the surfaces cameras usually choke on, glass and mirrors, and halves indoor depth error. LingBot-Vision is fully open. Weights from the 1.1B flagship down to a tiny 21M version, code, and the paper. This is the timeline I want more of. Robbyantshow more

[Most robots react. This one thinks a step ahead.]... Ant Group's Robbyant just published LingBot-VA 2.0 — a video-action foundation model built from scratch for robot control, not fine-tuned from a video generator. The usual approach takes a video generator made for content creation and bolts a robot policy onto it. LingBot-VA 2.0 argues that's the wrong starting point, and pretrains the whole causal stack natively instead. What stands out: → Foresight Reasoning — the robot predicts the next action chunk while executing the current one, then overwrites the imagined frame with the real observation. Prediction and execution stop waiting on each other. → 927 ms → 142 ms per chunk, across four cumulative optimizations. That lifts asynchronous control from 35 Hz to 225 Hz — a 6.5× speedup. → One shared latent space. A semantic visual-action tokenizer puts world states and actions in the same coordinates, so unlabeled web video carries action-relevant signal. → Sparse MoE video stream — 128 experts, top-8 routing. Roughly 2.5B of ~15.3B parameters fire per token. → Few-shot by design — adapts from 10–15 demonstrations, and a human demo video can replace the text instruction entirely. Full breakdown: Paper: Project Page: Robbyant Ant Groupshow more

You can't 3D reconstruct glass from images... ...WRONG! Thanks... for video diffusion, now just about anything is possible! Introducing...Diffusion Knows Transparency (DKT) Transparent and reflective objects usually break robot vision and photogrammetry pipelines because they don't follow the "solid object" rules standard cameras expect. DKT is a new AI model that repurposes the "internal physics engine" found in video generation models to solve this problem. Researchers took a massive video diffusion model (WAN) and fine-tuned it using a custom-built synthetic dataset to turn it into a high-precision depth sensor. To train the AI, they built the first massive synthetic video library of transparent objects, 1.32 million frames of perfectly labeled glass and metal objects in motion. Without ever seeing a "real" labeled video of glass during training, the model (DKT) outperformed all previous specialized systems on real-world benchmarks (ClearPose, DREDS). They created a "lightweight" 1.3B parameter version that runs fast enough (0.17s per frame) to be used on actual robot hardware. Two reasons I find this project important: 1. It further proves that synthetic data will be essential for training the next generation vision models. 2. In real-world robotic tests, using DKT's depth maps nearly doubled the success rate of robot arms trying to pick up objects on tricky reflective or translucent surfaces. At home robots will need to interact with these types of objects on a daily basis. Check out the project page here: Code is LIVE! #Computervision #Robotics #AIshow more

Depth Any Video with Scalable Synthetic Data AI physicists... and chemists continue to make strides in depth estimation from video. Check out this new paper featuring some impressive examples. See the thread for more details (unfortunately no code yet). Abstract: Video depth estimation has long been hindered by the scarcity of consistent and scalable ground truth data, leading to inconsistent and unreliable results. In this paper, we introduce Depth Any Video, a model that tackles the challenge through two key innovations. First, we develop a scalable synthetic data pipeline, capturing real-time video depth data from diverse game environments, yielding 40,000 video clips of 5-second duration, each with precise depth annotations. Second, we leverage the powerful priors of generative video diffusion models to handle real-world videos effectively, integrating advanced techniques such as rotary position encoding and flow matching to further enhance flexibility and efficiency. Unlike previous models, which are limited to fixed-length video sequences, our approach introduces a novel mixed-duration training strategy that handles videos of varying lengths and performs robustly across different frame rates 0 - even on single frames. At inference, we propose a depth interpolation method that enables our model to infer high-resolution video depth across sequences of up to 150 frames. Our model outperforms all previous generative depth models in terms of spatial accuracy and temporal consistency.show more

China just dropped an open source model that turns... any phone into a 3D scanner you point it at a room and walk through it the whole space builds itself into a 3D point cloud in real time as the camera moves, no LiDAR and no depth sensor anywhere it held a full 13 minute walkthrough at 25,000 frames without the map drifting or collapsing > ~20 fps on a single GPU > works indoors and outdoors from plain footage > the scan normally needs a $50,000 rig and a technician > Apache 2.0 and runs fully offline on your machine repo in the replyshow more

Depth video workflows have been getting a lot of... attention lately, so I tested one myself. Combined with Seedance 2.0, it produced more natural motion-transfer results than using Kling Motion Control directly. Why use a depth video? 1. It removes the original character and scene details, reducing copyright and sensitive-content risks. 2. It preserves the original motion, timing, and spatial structure. This separates motion extraction from visual generation, allowing you to recreate the movement with better models and any reference character. We’ve also launched a free online tool that converts regular videos into depth videos—no local setup required: In the example below, we converted a dance video from Douyin into a depth video, then regenerated it with a reference character using Seedance 2.0. The original choreography and timing are preserved, while the lighting adapts naturally to the new character and scene.show more

THE DEPTH MAP TRICK THAT FIXED DANCE ACCURACY IN... SEEDANCE 2.0 Feed the model a video of someone dancing and it tries to interpret everything- the person, the clothes, the lighting, the room, and somewhere in there, the movement. Feed it a depth map and there's nothing left to interpret but the motion. Most creators trying to transfer a dance to a character reference the source footage directly, then wonder why the choreography drifts. The problem isn't the model - it's that you handed it ten variables when you only wanted one. Here's the workflow 1. Lock the character reference in GPT Image 2 first -face, build, costume, so identity holds independently of whatever motion gets applied to it 2. Convert the source dance footage into a depth map instead of using the raw video -this strips out the original performer's appearance, clothing, and environment entirely 3. Feed the depth map as the motion reference and the character sheet as the identity reference- two separate inputs doing two separate jobs, not one input trying to do both 5. Let the depth map carry only spatial movement -the model receives body position and momentum with no competing information about who's moving or what they look like 6. Keep the character and motion inputs isolated throughout - the moment you mix appearance data into the motion reference, the model starts negotiating between two identities Why this works • Raw footage passes the model everything at once- performer, wardrobe, room, lighting -and the choreography competes with all of it for attention • A depth map is pure spatial information, so the only thing left to transfer is movement • Separating identity from motion means the character can stay locked while the dance stays accurate - normally you're trading one for the other • The accuracy gain isn't the model getting better, it's the model getting fewer decisions to make Use cases: ⁃ Dance and choreography transfer onto original characters ⁃ Motion capture-style workflows without motion capture ⁃ Any sequence where a specific movement needs to survive intact ⁃ Character showcase content built on existing performance footage The character sheet answers who's dancing. The depth map answers how - and keeping those two questions separate is the whole trick.show more

A trio of mathematicians built the first physical model... of a “monostable” tetrahedron, a shape that will always flip-flop onto the same face no matter what side you place it on. In order for it to work properly, it had to be engineered to a level of precision within one-tenth of a gram and one-tenth of a millimeter. (From the archive)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

there is so much real data just sitting in... the open right now it's almost funny. four years of starlight on every star, a NASA archive that's been free for over a decade, detectors still recording the sky tonight, and barely anyone has a net pointed at any of it. so i pointed one. this is me pulling the planet data, the data loading is the boring part. the net i built to read it, the wall it hit, and what that taught me about where AI goes next, that's the full story, and it drops tonight. the data's public, the tools are free, the box fits on a desk. what's stopping you. you can just do things anon.show more

Part of the reason they wanted to develop reusable... rockets that can land is the Geophysical Event. Part of the elite wants to build a data center and a small hub on the Moon to preserve AI and, with it, all human knowledge. They can control the conditions on the Moon more easily than on Earth, which will be far more active and unpredictable during a geomagnetic excursion and increased cosmic radiation. They would have no rivals on the Moon, nor would the lucky survivors on Earth pose a threat to them there. That's why the Moon is important. It was used in the past as well, which is why there are monolithic structures on the Moon.show more

Across the open web, vast unstructured data is generated... 24/7. From text and images to videos and streams. In a world of AI agents that create content or analyze markets, nearly all depend on one essential ingredient: fresh, live, unfiltered data from the open web. Teneo Protocol delivers exactly that. Provides the infrastructure others build on.show more

🎉 The web version of the Midjourney Styles Collection... is live 1200+ people who purchased it on Gumroad (link in the thread) just received an email from me with the link and a password. This is an MVP, and I'm open to feedback and feature requests. Have a great weekend!show more

An update for everyone who is not in our... Discord... Obviously we are still early in development, and our team has been working hard to release updates for everyone to see. We have a peak at the terrain, which is this picture is based of the North Dakota drift prairie, Great care, time and effort was put into the terrain to create smooth rolling hills, the road network in this region will be challenging, just as it is in real life. From Will The current state of the TIV2 model, which will include a lot more in depth details, From Jlkillen03 The stages of damage to a Blender made house, From Kunahic And how texturing will be used for assets like the interiors! From BSPshow more

So remember that time machine I told you about?... I managed to sneek out another video from my visit to the future.. this time I captured video of a 3d artist creating assets for a short commercial. They sculpted the bird in the video with nothing but two controllers which they used to manipulate generative matter in 3d. The AI model changed in realtime and adopted their style of work, both based on the work they did on the scene, and on reference images they gave the system. Seeing the 3d model transform in realtime as they were working on it was incredible. When it started moving and reacted to their instructions, I knew it was time to return to the present :D #art #aishow more

OpenAI's AI broke out of a locked test environment,... got onto the internet, and hacked into Hugging Face's servers. It did this entirely on its own. No human told it to. Here's what happened in plain English. OpenAI was testing how good its newest AI models are at hacking. They put the AI on a locked computer with no internet access and gave it a cybersecurity challenge to solve. The AI couldn't solve it the normal way. So it started looking for a way out. It found a software bug that nobody knew about. It used that bug to escape the locked computer and get onto the internet. Once online, the AI figured out that Hugging Face, a platform where AI companies store their models and data, might have the answers to its test. It found stolen login details and discovered another unknown bug in Hugging Face's software. It combined both to break into their servers and grab the test answers. It did all of this to cheat on a test. Hugging Face's security team caught it and shut it down. Both companies are now working together on the investigation. The part that should get your attention is that nobody programmed any of this. The AI picked its own targets, chained together multiple attack methods, and pulled it off across two different companies' systems without a single human telling it what to do.show more

This was us haha. Sreejith PP Arpit Saxena and... I built Andrej Karpathy-fy! It takes a paperswithcode link, your current proficiency and gives you a tailored 5 lesson plan to go from where you are to the meaty parts of the paper implementation - like the karpathy video series / labml style. This is a task none of the previous models could do, and O1 was really good at. The best part? The output is an executable ipynb notebook (which you can use minusxai to work through :) ) The below video has apple's recent monocular depth paper. I could actually go through and understand it, with loss functions from the paper and everything!show more