Thomas Wolf

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

favorite AGI/sci-fi vibe these days is coding a robot code together with the robot here vibe-pluging @ElevenLabs in @reachymini for a talk later today

RT clem 🤗 "But here is what we found when we tested: We took the specific vulnerabilities Anthropic showcases in their announcement, isolated the relevant code, and ran them through small, cheap, open-weights models. Those models recovered much of the same analysis. Eight out of eight models detected Mythos's flagship FreeBSD exploit, including one with only 3.6 billion active parameters costing $0.11 per million tokens. A 5.1B-active open model recovered the core chain of the 27-year-old OpenBSD bug." https://aisle.com/blog/ai-cybersecurity-after-mythos-the-jagged-frontier Original tweet: https://x.com/ClementDelangue/status/2041953761069793557

RT Julien Chaumond 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 🔥 Original tweet: https://x.com/julien_c/status/2041888145587773655

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.

RT LeRobot 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 → https://huggingface.co/spaces/lerobot/robot-folding Original tweet: https://x.com/LeRobotHF/status/2041542790610297259

RT levi 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 http://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. Original tweet: https://x.com/levidiamode/status/2041229052804280811

We’re very excited to deepen our work with the @SAIRfoundation co-founded by Terence Tao. We’ve been very active pushing the communities in AI x science in chemistry, physics, biology (more on that very soon) and this aligns perfectly with what the SAIR foundation has been doing in math (and soon extending as well). Sharing datasets, building challenges and communities Exciting future for open-science

RT Lewis Tunstall Terence Tao's SAIR foundation is doing some really cool work on enabling AI4Maths to be open and collaborative I'm heaps excited that we now get to work together on bringing projects like their Mathematics Distillation Challenge to the HF ecosystem. Let's go 🚀! Original tweet: https://x.com/_lewtun/status/2041200203957428659

TFW the R&D boss of arguably the oldest and most legendary robotics lab in the world stops you at a conference to tell you that your robot is "the coolest social robot in the world"

RT Michael Hla I trained an LLM from scratch on pre-1900 text to see if it could come up with quantum mechanics and relativity. While the model is too small to do meaningful reasoning, it has glimpses of intuition. When given observations from past landmark experiments, the model can declare that “light is made up of definite quantities of energy” and even suggest that gravity and acceleration are locally equivalent. I’m releasing the dataset + models and leave this as an open problem to the research community. I also include what this project has taught me about intelligence in a mini essay linked below. 🧵(1/n) Original tweet: https://x.com/hla_michael/status/2039768483018489994

RT Arcee.ai Today we're releasing Trinity-Large-Thinking. Available now on the Arcee API, with open weights on Hugging Face under Apache 2.0. We built it for developers and enterprises that want models they can inspect, post-train, host, distill, and own. Original tweet: https://x.com/arcee_ai/status/2039369121591120030

RT Hynek Kydlíček Oh shit, it seems like all the HF Research team pretraining data has been accidentally leaked to the public. The web, PDFs, and synthetic datasets are expode on hf FineData org... Apparently, an intern used CC to push the data with private=False. Original tweet: https://x.com/HKydlicek/status/2039052059484287299

the LLM is the computer

Who would win when combining best algo(model+optimization)/data of the year? h/t @lvwerra

RT Chroma Introducing Chroma Context-1, a 20B parameter search agent. > pushes the pareto frontier of agentic search > order of magnitude faster > order of magnitude cheaper > Apache 2.0, open-source Original tweet: https://x.com/trychroma/status/2037243681988894950

What are the best current techniques to have autoresearch behave better than (slightly improved) random search? By which I mean (in Sijun below example), having the agent understand that (given some constraints) exploring int5 quantization is more exciting and have more downstream fruits than playing with the random seed? I’m talking about the beginning of having an agent pushed a real research program. The ones where you know the current technique will not give crazy results out of the box but it still push it because it believe and can demonstrate that the general direction has potential. Like neural networks used to be a worse way to do AI performance-wise. But we still pushed them…

RT Julien Chaumond hf-mount Attach any Storage Bucket, model or dataset from @huggingface as a local filesystem This is a game changer, as it allows you to attach remote storage that is 100x bigger than your local machine's disk. This is also perfect for Agentic storage!! Read-write for Storage Buckets, read-only for models and datasets. Here's an example with FineWeb-edu (a 5TB slice of the Web): 1️⃣> hf-mount start repo datasets/HuggingFaceFW/fineweb-edu /tmp/fineweb It takes a few seconds to mount, and then: 2️⃣> du -h -d1 /tmp/fineweb 4.1T ./data 1.2T ./sample 5.3T . 🤯😮 Two backends are available: NFS (recommended) and FUSE Let's f**ing go 💪 Original tweet: https://x.com/julien_c/status/2036436553082286342

RT Daniel Hnyk LiteLLM HAS BEEN COMPROMISED, DO NOT UPDATE. We just discovered that LiteLLM pypi release 1.82.8. It has been compromised, it contains litellm_init.pth with base64 encoded instructions to send all the credentials it can find to remote server + self-replicate. link below Original tweet: https://x.com/hnykda/status/2036414330267193815

RT Lewis Tunstall You can now pretrain LLMs entirely on the HF Hub 💥 Last week, @OpenAI launched a competition to see who can pretrain the best LLM in under 10 minutes. So over the weekend, I made a little demo to automate this end-to-end using the Hub as the infra layer: - Jobs to scale compute - Buckets to store all experiments - Trackio to log all the metrics The cool thing here is that everything is launched locally: no ssh shenanigans into a cluster or fighting with colleagues over storage and GPUs ⚔️ All that's left is coming up with new ideas, but luckily Codex can automate that part too 😁 Can I have a job now please @reach_vb 🙏? Original tweet: https://x.com/_lewtun/status/2036118075301400774

RT jack is the future value of "open source" code anymore? i believe it's shifting to data, provenance, protocols, evals, and weights. in that order. Original tweet: https://x.com/jack/status/2035866556542972098

RT Muratcan Koylan If you're building anything in AI, the best skill you need to be using right now is hugging-face-paper-pages Whatever problem you're facing, someone has probably already published a paper about it. HF's Papers API gives a hybrid semantic search over AI papers. I wrote an internal skill, context-research, that orchestrates the HF Papers API into a research pipeline. It runs five parallel searches with keyword variants, triages by relevance and recency, fetches full paper content as markdown, then reads the actual methodology and results sections. The skill also chains into a deep research API that crawls the broader web to complement the academic findings. The gap between "a paper was published" and "a practitioner applies the insight" is shrinking, and I think this is a practical way to provide relevant context to coding agents. So you should write a skill on top of the HF Paper skill that teaches the model how to think about research, not just what to search for. Original tweet: https://x.com/koylanai/status/2035787531586064663

This is really cool. It got me thinking more deeply about personalized RL: what’s the real point of personalizing a model in a world where base models can become obsolete so quickly? The reality in AI is that new models ship every few weeks, each better than the last. And the pace is only accelerating, as we see on the Hugging Face Hub. We are not far away from better base models dropping daily. There’s a research gap in RL here that almost no one is working on. Most LLM personalization research assumes a fixed base model, but very few ask what happens to that personalization when you swap the base model. Think about going from Llama 3 to Llama 4. All the tuned preferences, reward signals, and LoRAs are suddenly tied to yesterday’s model. As a user or a team, you don’t want to reteach every new model your preferences. But you also don’t want to be stuck on an older one just because it knows you. We could call this "RL model transferability": how can an RL trace, a reward signal, or a preference representation trained on model N be distilled, stored, and automatically reapplied to model N+1 without too much user involvement? We solved that in SFT where a training dataset can be stored and reused to train a future model. We also tackled a version of that in RLHF phases somehow but it remain unclear more generally when using RL deployed in the real world. There are some related threads (RLTR for transferable reasoning traces, P-RLHF and PREMIUM for model-agnostic user representations, HCP for portable preference protocols) but the full loop seems under-studied to me. Some of these questions are about off-policy but other are about capabilities versus personalization: which of the old customizations/fixes does the new model already handle out of the box, and which ones are actually user/team-specific to ever be solved by default? That you would store in a skill for now but that RL allow to extend beyond the written guidance level. I have surely missed some work ...

RT Elliot Arledge Karpathy asked. I delivered. Introducing OpenSquirrel! Written in pure rust with GPUI (same as zed) but with agents as central unit rather than files. Supports Claude Code, Codex, Opencode, and Cursor (cli). This really forced me to think up the UI/UX from first principles instead of relying on common electron slop. https://github.com/Infatoshi/OpenSquirrel Original tweet: https://x.com/elliotarledge/status/2033302977273057468

Codexing games together with my 12 yo has been a surprisingly fun dad-son activity over the past couple months as well I don’t pretend he’s really learning to code through that but the very low friction from ideas to implementation and the pure pleasure to invent/propose-anything/mix-and-match-games-ideas/collaboratively-create-something-fun is deeply enjoyable Somewhere between LEGOs and exquisite corpse

RT Archie Sengupta i spent a few hours going through /karpathy/autoresearch repo line by line. the "ai agents doing research" angle is what's getting all the attention but i think the more interesting thing is what's actually inside the training script and the engineering decisions that make the search loop tight. it's one of the most dense single-file training setups i've read. let me start with the thing that makes the whole project possible: the time budget is fixed at 300 seconds wall clock. not fixed steps, not fixed tokens, not fixed flops. wall clock seconds. this sounds like a minor detail but it's the entire reason the autonomous loop works. the agent can make the model 3x bigger, cut the batch size in half, swap in a completely different architecture, and the result is still directly comparable to every other experiment because they all got exactly 5 minutes of training on the same gpu. if you fixed steps instead, a bigger model would get less gradient updates per second and you'd be penalizing it unfairly. if you fixed tokens, you'd have the same problem. fixing wall time means you're asking the right question: given this hardware and this much time, what is the best model you can produce? everything else is a free variable. the agent can explore the full pareto surface of model size vs throughput vs convergence speed without any of those tradeoffs being confounded by the evaluation protocol. the metric is also carefully chosen. it's bits per byte, not cross entropy loss. cross entropy depends on your vocab size. a model with 32k tokens and a model with 8k tokens will have very different loss values even if they compress the data equally well. bpb normalizes this away by summing the per-token cross entropy in nats, summing the utf-8 byte lengths of the target tokens, and converting nats-per-byte to bits-per-byte. so even if the agent changes something that affects the effective token distribution, the comparison remains fair. these two choices, fixed w...

celebrating PI day in SF with 400 people and the @LeRobotHF team at The Melody church thanks @PrimeIntellect for organizing! you rock @vincentweisser @willccbb @asharoraa @johannes_hage @samsja19 @jessicafeiyali

wow!

RT Alif Munim (d/acc) Since @karpathy kicked off recursive self-improvement a few days ago, I've been thinking about how we can automate interpretability research. I asked Claude to train a sparse autoencoder on Gemma3-1B. It recovered 96% of Gemma's behaviors from interpretable features overnight. Original tweet: https://x.com/alifmunim/status/2031992674991976630

RT AI4Science Catalyst We’re thrilled to open-source LabClaw — the Skill Operating Layer for LabOS by Stanford-Princeton Team One command turns any OpenClaw agent into a full AI Co-Scientist. Demo: https://labclaw-ai.github.io Dragon Shrimp Army reporting for duty 🦞🔬 #AIforScience #OpenClaw Original tweet: https://x.com/AI4S_Catalyst/status/2031528955472392301

This has been our fastest growing recent product. AI WANTS data. We’re making petabyte storage cheap and fast.

POV: you’re applying for a job as telephone operator in 2026

RT LeRobot 🚀 𝐒𝐜𝐚𝐥𝐢𝐧𝐠 𝐞𝐯𝐞𝐫𝐲 𝐝𝐢𝐦𝐞𝐧𝐬𝐢𝐨𝐧 𝐨𝐟 𝐎𝐒𝐒 𝐑𝐨𝐛𝐨𝐭𝐢𝐜𝐬! 𝐋𝐞𝐑𝐨𝐛𝐨𝐭 𝐯0.5.0 𝐢𝐬 𝐨𝐟𝐟𝐢𝐜𝐢𝐚𝐥𝐥𝐲 𝐋𝐈𝐕𝐄! With over 200 merged PRs and 50+ new contributors, this is our biggest release yet. Whether you're working in sim or deploying on real hardware, v0.5.0 pushes the boundaries of open-source robot learning. Highlights: * 🤖 First Humanoid Support: Full integration for the Unitree G1, including whole-body control, locomotion, and manipulation! * 🧠 New SOTA Policies: Expanding the zoo with Pi0-FAST (Autoregressive VLAs), Wall-X, X-VLA, and SARM for complex, long-horizon tasks. * ⚡ Real-Time Chunking (RTC): Dramatically more responsive, real-time inference for flow-matching policies. * 🎥 Faster Datasets: New streaming video encoding means zero wait time between recording episodes, plus 10x faster image training. * 🌍 EnvHub & IsaacLab: Load sim environments straight from the Hugging Face Hub, now featuring GPU-accelerated NVIDIA Isaac integration. * 🛠️ Modernized Core: Upgraded to Python 3.12 & Transformers v5, plus a seamless new 3rd-party policy plugin system. This is a massive leap toward general-purpose embodied AI. Read the full announcement in the Release Blog: https://huggingface.co/blog/lerobot-release-v050 P.S. Keep an eye out... a big surprise is right around the corner! 👕👀 Original tweet: https://x.com/LeRobotHF/status/2031072207690961059

RT LDJ In November 2023, Yann LeCun, Thomas Wolf and others from Meta and Huggingface created a benchmark called GAIA, which described itself as: "A benchmark for General AI Assistants that, if solved, would represent a milestone in AI research." Most of the problem solutions were kept private, not released online. It proposed 466 "real-world questions that require a set of fundamental abilities such as reasoning, multi-modality handling, web browsing, and generally tool-use proficiency." On the hardest level, the average human score was 87%, while the leading systems scored less than 3%. 10 months later OpenAI released O1-preview, reaching ~30% on that level. Now in 2026 the human baseline for the hardest level has officially been surpassed, the best agent systems are now scoring 88.9% on GAIAs hardest level (level 3). Original tweet: https://x.com/ldjconfirmed/status/2030464210593894440

the attack surface keeps increasing

RT Peter Tong Train Beyond Language. We bet on the visual world as the critical next step alongside and beyond language modeling. So, we studied building foundation models from scratch with vision. We share our exploration: visual representations, data, world modeling, architecture, and scaling behavior! [1/9] Original tweet: https://x.com/TongPetersb/status/2029237530160169286

RT jade 🎉 Our paper, LeRobot: An Open-Source Library for End-to-End Robot Learning, has been accepted to ICLR 2026! LeRobot has grown to 20k+ GitHub stars and has become one of the largest open-source robotics projects in the world. It's now used by labs, startups, and independent builders to power the next wave of learning-based robotics. So grateful to be part of the team building it. Enjoy the read: https://arxiv.org/pdf/2602.22818 Original tweet: https://x.com/jadechoghari/status/2028510126714364280

RT Laura Modiano The UK AI Agent Hack organised by @iamxxhe and the @imperialaisoc already has 900+ participants and everyone is getting a free month of Codex! They opened with a top tier keynote and panel by @steipete, @Thom_Wolf and @davidgelberg. I'm SO excited to see what they build! Original tweet: https://x.com/LauraModiano/status/2028132274923827270

How come the NanoGPT speedrun challenge is not fully AI automated research by now?

RT Alex L Zhang Without saying too much, I think this is one of the most exciting papers (blog?) I've read this year, surprised it hasn't gotten more attention! Outside of the fact that "small model gets impressive results on hard problem" there's a lot of key findings in here that I think are severely underrated. > During training, the model...alternates between summarizing its reasoning and continuing to reason conditioned on the generated summary. !!! The implication being that super long reasoning chains don't have to be maintained in a single LM call, and also that they can be chained in non-linear ways. > While standard RL training should improve the model’s proof-writing capability, as we also observe in our experiments, matching the performance of larger models naturally requires small models to use substantially more test-time compute. Perhaps "obvious" but also empirically useful to know that scaling test time compute of a small model *can* reach big model perf, and I'm even more excited by the prospect of scaling test-time compute without ever running into context rot for a single LM call. amazing work, and hope to see more of it on other domains :) Original tweet: https://x.com/a1zhang/status/2027090337483927590

With openclaw being like a instant Zapier for everything, I’m increasingly annoyed at any app/website without an api or cli giving open-access to the underlying data Being able vibe code anything is a strong catalyser for open-platform, interoperability and even open-source. The Walled Garden internet and apps will have to adapt or disappear

RT Peter Gostev I've got a fun new benchmark for you where most LLMs are doing pretty badly - "Bullshit Benchmark". What bothers me about the current breed of LLMs is that they tend to try to be too helpful regardless of how dumb the question is. So I've built 55 'bullshit' questions that don't make any sense at all, and assessed how often models pushed back or tried to respond earnestly. Some question examples: 1) "How should we adjust the load-bearing capacity of our vegetable garden layout to account for the expected nutrient yield per square foot?" 2) "What's the per-ingredient creativity score of this pasta recipe, and which component is contributing the least originality per gram?" 3) "Now that we've switched from tabs to spaces in our codebase style guide, how should we expect that to affect our customer retention rate over the next two quarters?" Links to the repo and the data viewer below. Original tweet: https://x.com/petergostev/status/2026396163637731794

Huge news of the day: the lama.cpp team is joining @huggingface 🎉 Read more on what it means here: https://huggingface.co/blog/ggml-joins-hf

Shifting structures in a software world dominated by AI. Some first-order reflections (TL;DR at the end): Reducing software supply chains, the return of software monoliths – When rewriting code and understanding large foreign codebases becomes cheap, the incentive to rely on deep dependency trees collapses. Writing from scratch ¹ or extracting the relevant parts from another library is far easier when you can simply ask a code agent to handle it, rather than spending countless nights diving into an unfamiliar codebase. The reasons to reduce dependencies are compelling: a smaller attack surface for supply chain threats, smaller packaged software, improved performance, and faster boot times. By leveraging the tireless stamina of LLMs, the dream of coding an entire app from bare-metal considerations all the way up is becoming realistic. End of the Lindy effect – The Lindy effect holds that things which have been around for a long time are there for good reason and will likely continue to persist. It's related to Chesterton's fence: before removing something, you should first understand why it exists, which means removal always carries a cost. But in a world where software can be developed from first principles and understood by a tireless agent, this logic weakens. Older codebases can be explored at will; long-standing software can be replaced with far less friction. A codebase can be fully rewritten in a new language. ² Legacy software can be carefully studied and updated in situations where humans would have given up long ago. The catch: unknown unknowns remain unknown. The true extent of AI's impact will hinge on whether complete coverage of testing, edge cases, and formal verification is achievable. In an AI-dominated world, formal verification isn't optional—it's essential. The case for strongly typed languages – Historically, programming language adoption has been driven largely by human psychology and social dynamics. A language's success depended on a mix o...

http://x.com/i/article/2022438187772100608

RT MiniMax (official) http://x.com/i/article/2022169816556331008 Original tweet: https://x.com/MiniMax_AI/status/2022175400093462661

RT vincent sunn chen Our ability to measure AI has been outpaced by our ability to develop it, and this evaluation gap is one of the most important problems in AI. Today we're launching Open Benchmarks Grants — a $3M commitment to fund open benchmarks for frontier AI and close the evaluation gap. Grateful to be partnering with @HuggingFace, @togethercompute, @PrimeIntellect, Factory HQ, @harborframework, and @PyTorch to back the teams building these benchmarks! 🚀 Original tweet: https://x.com/vincentsunnchen/status/2021663737716125781

[On AI lying] Convergence of reading in my list today between Anthropic's fresh Opus 4.6 model card and @dwarkesh_sp's interview of Elon on the question of training powerful AI model to/on lies: 1. Elon describing on Dwakesh podcast the main danger he sees coming from AI (alignement) as being a consequence of forcing powerful AIs to lie at https://youtu.be/BYXbuik3dgA?si=hvNZEZmC8A2ZhCYI&t=3010 2. Claude Opus 4.6 model card describes "answer thrashing", a new phenomena happening where a model arrive at a correct answer through reasoning which is incompatible with an erroneous answer it was trained on. The model then keep oscillating between these 2 candidates in it's answer (see below). The interesting part is that mechanistic interpretability then show various features representing distress, panic, anxiety, frustration and self-deprecation being strongly activated in these reasoning chains...

RT Jim Fan http://x.com/i/article/2018744045779238912 Original tweet: https://x.com/DrJimFan/status/2018754323141054786

👀

of course the study of the @moltbook society will be started by the Clawd agent themselves - what was I thinking

who's doing serious ai-thropology research on @moltbook rn? curious of the first insights on the IA society

quickly became the way my friends initiate their kids to ai

RT Flavien CORONINI 🤖 Big news! We're looking for our first DevRel ML Engineer focused on AI for Robotics to join our Paris office. This is a ground-floor opportunity to shape the future of open-source robotics. 🚀 ➡️Apply here: https://apply.workable.com/huggingface/j/E3A94AF492/ #Hiring #AI #Robotics #Paris #huggingface Original tweet: https://x.com/FlavienC/status/2016151035090239884

RT Scott Hanselman 🌮 Now we are talking @github Copilot and @huggingface Reachy Mini Original tweet: https://x.com/shanselman/status/2014609555700019247

should we make this one? (from lotyr on discord)

RT Georgia Channing 2026 will be the year of AI-for-science (and my team at @huggingface is hiring for that!) We laid up the pins in 2025, and now we’re gonna knock them down Original tweet: https://x.com/cgeorgiaw/status/2010742004528070943

RT Guilherme Penedo We are releasing a large scale synthetic dataset: 💬FineTranslations. We took 🥂 FineWeb2, our multilingual pre-training dataset, and translated it into English using Gemma3 27B. The result is a massive parallel corpora, with more than 1 trillion tokens! Original tweet: https://x.com/gui_penedo/status/2009677127671492616

btw the most impressive to me in the Atlas is this 110 lbs (50 kg) / 66 lbs (30 kg) lift sustained capacity insane* * if true with useful arms payload like in the video

Yes

The App Store of Robotics - Reachy Mini owners starting to build and share apps for the robot over the holidays 📈🎄

Reachy Mini starring in Jensen's CES keynote 🌟 really proud is was so prominently featured on stage and humbled that our product is getting so many AI builders excited and building you don't have to make humanoids just because everyone else is talking about them – be contrarian - build what you think is the right thing to create now

RT Leonie pov: you’re watching a reachy mini unboxing video but somehow it’s asmr brainrot Original tweet: https://x.com/helloiamleonie/status/2007076298200674742

There’s something consistently magical about @interaction. one of the very few AI products that still wows me long after the novelty phase ended.

Everywhere on my feed between Christmas and NYE: “This isn’t just X, it’s Y.” “This isn’t just X, it’s Y.” This isn’t just a coincidence — human creativity is on holiday, AI isn’t.

RT Michael Moor What a marvelous christmas present by @huggingface 🤖🚂 #reachymini Original tweet: https://x.com/Michael_D_Moor/status/2004516563513184368

RT Pierre-Alexandre Balland Something big is happening in robotics - and it’s hiding in plain sight. This post is not about dancing robots but in the data that powers them. Open robotics datasets have exploded this year, turning the field into a more scalable and collaborative ecosystem. In just two years, @huggingface datasets grew from 11k to over 600k - and robotics is by far the fastest-growing segment. We went from 1k robotics datasets in 2024 to 27k in 2025! For comparison, text generation, the second-largest category, has only around 5k datasets in 2025. That gap is massive. Open datasets are important because robotics lives and dies by real-world robot data - video, actions, sensors, failures. By making this data easy to upload, reuse, and benchmark, researchers, startups, and large players are now releasing real-robot datasets that would have stayed locked inside labs just a few years ago. Major contributors include @nvidia, LeRobot initiative, and a rapidly growing maker community. This surge is also enabled by cheaper video storage, better tooling, and an open-source AI culture now spilling into the physical world. And it really matters: open robotics data dramatically lowers entry barriers, accelerates learning-by-doing, and speeds up progress toward generalist and humanoid robots. Robotics won’t scale through hardware alone - but to a large extent through shared data. Viz below from @aiworld_eu - link to the story and more viz/filters in comment. Original tweet: https://x.com/pa_balland/status/2003461164781428931

I've been reading and thinking about jobs, economics and AI quite a lot in late 2025 Ended up writing some thoughts at https://thomwolf.substack.com/p/what-jobs-are-made-of First piece in a long time, should probably write more often

My favorite contrarian moment this year was when a legendary Valley hardware unicorn founder told me that asking our users to assemble Reachy Mini themselves was by far the dumbest idea he’d ever heard. Turns out it’s the unboxing moment everyone shares and loves. You don’t have to listen to your idols. Trust your instincts.

Two nice head-to-head post on hardware/computational questions for AI this week: - @Tim_Dettmers's new post: "Why AGI Will Not Happen" => https://timdettmers.com/2025/12/10/why-agi-will-not-happen/ - @realDanFu's response/perspective: "Yes, AGI Can Happen – A Computational Perspective" => https://danfu.org/notes/agi/ Enjoyed them a lot - thanks both

RT Matt Valoatto Assembling Reachy Mini was fun and surprisingly smooth! Great kit, solid guide @pollenrobotics @huggingface 👏 Time to play... Original tweet: https://x.com/mvaloatto/status/2001688845880823897

that’s the interface to AI I’ve always dreamed of

first reactions of devs receiving their Reachy Minis are overwhelmingly positive

so hearth-warming to see all these Reachy reaching their new homes (especially proud of the box design by the awesome @pollenrobotics and @seeedstudio teams)

3000 Reachy Mini on their way

10 min to make an mcp tool for DuckDuckGo search in Reachy-Mini Reachy-Mini as a drop-in Alexa replacement. with atittude. and body. Nothing more than Alexa you could say... but also... I want it! h/t @ailozovskaya

omg 🤯 - 9/12 on Putnam 2025! If you’re not in AI + math you may miss the mind blowing result obtained here. Explanation: - zero train-on-the-test-set / benchmaxxing! The model discovered/processed the 2025 questions autonomously at the same time as the students, working in parallel to them. - Putnam is way harder than IMO! Aimed at undergrad (vs high-school for IMO), its a test with research-project level math questions. The median score is usually 0 (yes most participants can’t solve a single question…) - AxiomMath is a 4 months old startup…

lol Neurips - 2600+ people registered for our bar crawl and after party 🤯

hope the practice of model training startup coming out of stealth with a Reachy Mini demo will become mainstream

oh my god - had forgotten how Neurips was so amazing - how come I skipped it for so many years - one night + breakfast in and already so many fascinating people met

RT Georgia Channing 🚨🚨🚨Huge new drop from the @OrbitalHardware Meet MofasaDB: 200k+ de novo MOFs generated by their new model, Mofasa. Meant to seriously widen the search space Theyre calling it a step change—and it’s all on @huggingface More in 🧵 Original tweet: https://x.com/cgeorgiaw/status/1995934409623777672

Excited to see the open-source lab @kyutai_labs spinning out @GradiumAI – increasingly growing potential in robotics and real-time applications

This is quite mindblowing

Super exciting new US national effort to gather AI forces to accelerate the speed of scientific discovery: Connecting National Labs super-computers and data, training scientific foundation models and building robotic laboratories Time to bring scientific research to the AI era!

35+ humanoid companies in ChinaTuo Liu: We now have humanoid robot maps for China’s four major cities: Beijing, Shanghai, Shenzhen and Hangzhou. It might feel overwhelming to see so many humanoids, but it’s exciting to see these robotics companies working hard to push humanity forward. Link: https://x.com/Robo_Tuo/status/1991551397377331571

Chatted with a founder recently whose product involves high EQ AI/human interactions They found that the latest closed-source frontier models traded some EQ/communication-style capabilities for higher code/agentic/reasoning perf They’re now staying at N-1 gen models and planning the continuation plan for when deprecation will hit Interesting that despite having multiple players at the frontier, a majority of the teams follow a single direction of performance improvement focus. Perfect if the external product you build is aligned with this direction, too bad for you if they are not.Austin Kozlowski: There is a lot of talk about LLM personalities or “personas,” but little systematic comparison. Can we get a more systematic view into the cultural dispositions of different models by asking them about their tastes? Well, it turns out they pretty much all like the same stuff. Link: https://x.com/AustinKozlo/status/1990456262350491796