AI Dynamics

Global AI News Aggregator

About

@thom_wolf

  • Team of agents solving a physics problem

    By

    @thom_wolf

    –

    Watching a team of agents tackle a challenging theoretical physics problem is quite mesmerizing—they self-correct, derive complex equations, compute intermediate results, and re-evaluate the best approach.

    → View original post on X — @thom_wolf

  • Science Resources Hub: Chemistry Biology Physics Materials Math

    By

    @thom_wolf

    –

    Super happy to have this one out. A clean organized up-to-date view of all the science resources (chemistry, biology, physics, materials, math) people have been sharing on the Hugging Face hub: datasets, blogs, models and more

    → View original post on X — @thom_wolf

  • AI Agents Challenge GitHub’s Centrality in Code Distribution

    By

    @thom_wolf

    –

    GitHub central place might become challenged in a world where (1) we access/get code and libraries through agents/chats and (2) our codebases are increasingly custom tailored and build from scratch Agents explore the web better than us and can get/store code from many places

    → View original post on X — @thom_wolf

  • Hugging Face Ecosystem Powers SOTA ML Training Agents

    By

    @thom_wolf

    –

    Love this work from Aksel and the post-training team at Hugging Face! Turns out the HF ecosystem (papers, datasets, models all accessible through CLI, skills and md files) is perfect for running SOTA ML agents: agents that can train any type of AI model to top performance. A

    → View original post on X — @thom_wolf

  • AsyncGRPO TRL Library: Scaling Inference Training Decoupling
    AsyncGRPO TRL Library: Scaling Inference Training Decoupling

    By

    @thom_wolf

    –

    **Deep content post alert** A technical deep dive for your Sunday morning, somewhere between a short detective story and a tutorial on RLHF We recently added AsyncGRPO in the TRL library to decouple inference and training and scale much faster and harder. As a sanity

    → View original post on X — @thom_wolf

  • Great Release and Results Announcement

    By

    @thom_wolf

    –

    Great to see this released (and results)

    → View original post on X — @thom_wolf

  • JSON’s Token Inefficiency Problem Costs Real Money

    By

    @thom_wolf

    –

    json is so token inefficient it hurts these days man, these braces and quotes are costing me real $$

    → View original post on X — @thom_wolf

  • PyTorch Foundation Receives Safetensors from Hugging Face

    By

    @thom_wolf

    –

    We are giving away Safetensors to the @pytorch foundation (shepherded by the Linux Foundation) Our shared goal is to make the default serialization format for torch safe and performant. To unlock this, governance needs to be independent of @huggingface. Looking forward to more stakeholders contributing to Safetensors in the coming months 🔥

    → View original post on X — @thom_wolf

  • LeRobot Releases Open-Source Bimanual Clothes Folding Robot

    By

    @thom_wolf

    –

    Releasing one of our *largest* robotics project yet in the open We collected and annotated hours of clothes folding with open-arms and collaborators. We then explored how to train the best clothes folding robotic model for bimanual setups. And now we're releasing it all fully in the open: data, code, models, software, explorations, learnings, you name it Enjoy, play with it, use these learnings and share yours! PS: the hub is increasingly *the* place where robotics data is being shared and used, come take a look if you haven't yet. Robotics data has been our fastest growing dataset category by far over the past few months. LeRobot (@LeRobotHF) Releasing the Unfolding Robotics blog! Time to unfold robotics: we trained a robot to fold clothes using 8 bimanual setups, 100+ hours of demonstrations, and 5k+ GPU hours. Flashy robot demos are everywhere. But you rarely see the real story: the data, the failures, the engineering. We’re sharing everything: code, data, and details in the blog → huggingface.co/spaces/lerobo… — https://nitter.net/LeRobotHF/status/2041542790610297259#m

    → View original post on X — @thom_wolf

  • The Ultra-Scale Playbook: GPU Parallelism Strategies for LLM Training
    The Ultra-Scale Playbook: GPU Parallelism Strategies for LLM Training

    By

    @thom_wolf

    –

    Day 93/365 of GPU Programming Studying parallelism today and stumbled upon this incredible blog post/book The Ultra-Scale Playbook: Training LLMs on GPU Clusters by Hugging Face that dives deep into data parallelism, expert parallelism, tensor parallelism, pipeline parallelism and context parallelism. I've read a bit about each of these methodologies before but this is the best resource I've found that really pieces them all together into a unified coherent picture. Kinda like its name implies, the team goes into actual empirical examples based on the 4000 scaling experiments (across up to 512 GPUs!) they conducted. E.g. how does tensor parallelism reduce activation memory for matmuls but still require gathering full activations for LayerNorm? When does pipeline parallelism's bubble overhead outweigh its memory savings? When and why would you combine TP/PP/DP on a specific cluster topology? What's the real memory breakdown between params, gradients, optimizer states and activations and which parallelism strategy targets which? et cetera Also loved all the beautiful and sometimes interactive diagrams that reminded me of distill.pub (which makes sense given they used distill's template to create the post). I wish more blog posts in ML would use a similar approach to help visual learners understand the content at an intuitive level. Especially now that rich visualizations/animations are so easy to spin up with LLMs. Really wonderful work by @Nouamanetazi @FerdinandMom @xariusrke @mekkcyber @lvwerra @Thom_Wolf. In times when things are going more and more closed source in, this is such a good example of what great open source AI education and research can look like. levi (@levidiamode) Day 92/365 of GPU Programming Taking a closer look at disaggregated LLM inference today, which I've been wanting to survey more after listening to the Dean <> Daly discussion at GTC. The best resource I found on the topic was this great talk by @Junda_Chen_ on the past, present and future of prefill decode disaggregation. In the lecture, Junda goes through Nvidia's dynamo, the intrinsic tradeoff spectrum between throughput & latency, TTFT, TPOT, the "goodput" metric, distinct characteristics between prefill vs decode, chunking P&D, the problem of interference, pipeline parallelism, resource & parallelism coupling, disaggregation and DistServe. — https://nitter.net/levidiamode/status/2040938107604742640#m

    → View original post on X — @thom_wolf, 2026-04-06 18:58 UTC