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

OpenAI's Deep Research is getting a run for its... money. Deep Lake was just released, and it's a different take on an AI system that can do deep research on your own data. You can use Deep Lake to build AI search with reasoning on your private and public data. (Look at the attached videos to get an idea of how it works.) If you want to research proprietary and sensitive data, Deep Research won't help you because it's limited to public data. Deep Lake, however, will allow you to use your private data. On top of that, Deep Lake supports multi-modal retrieval from the ground up. It uses vision language models for data ingestion and retrieval so that you can connect any data (PDFs, images, videos, structured data, etc.) You can even use mixed-data queries! Deep Lake can search your data from S3, Dropbox, and GCP. It learns from your queries over time, making the results as relevant to your work as possible!show more

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

🚨 JUST IN: CHINA just released an AI EMPLOYEE... that works 24X7 on its own. 100% OPEN SOURCE. It researches, codes, builds websites, creates slide decks, and generates videos. All by itself. All on your computer. It's called DeerFlow. You give it a task. It makes a plan, spins up its own team of sub-agents, and gets to work. You come back and there's a finished deliverable waiting. Not a draft. Not a summary. The actual thing. Not a chatbot. Not a research assistant. An AI with its own computer that works while you sleep. Here's what it does on its own: → Spawns multiple sub-agents in parallel, each tackling a different piece of your task, then combines everything into one finished output → Writes real code, runs it, reads the results, and fixes its own mistakes without asking you once → Builds slide decks, websites, full research reports, and data dashboards from scratch → Remembers you across sessions. Your writing style. Your tech stack. Your preferences. Gets better every time. → Reads files you upload, works with them inside its own filesystem, hands you clean finished outputs → Searches the web, runs commands, calls any tool you plug in Here's how it thinks: You give one instruction. The lead agent makes a plan. Sub-agents fan out and work in parallel. Results come back. Everything gets synthesized. You get a deliverable. A single research task might split into a dozen sub-agents, each exploring a different angle, then converge into one finished website with generated visuals. Here's the wildest part: DeerFlow 2.0 launched on February 28th 2026 and hit number 1 on all of GitHub Trending the same day. Version 2.0 was a complete rewrite. Zero shared code with version 1. Because users kept using it for things the team never intended. Data pipelines. Dashboards. Entire content workflows. The community told them what it needed to become. So they burned it down and rebuilt it. 22.7K GitHub stars. 2.7K forks. Built by ByteDance 100% Open Source. MIT License.show more

3 weeks since ml-intern launched and we just hit... 1M messages exchanged. that's 3.3 agent-years of ML research in 21 days. 2 months worth of research every day. 17,383 training jobs total. talk about AI acceleration. here's some of what people built: Carlos Miguel Patiño replicated the full DeepSeek v4 architecture and pre+post trained a 100M MoE from scratch. → it landed a third place submission on Keller Jordan optimizer competition. autoresearch on SOTA territory. Lewis Tunstall Got the intern to convert Alec Radford's cool new talkie-lm 1930 model to work with transformers. tokenizer, chat template, model conversion etc all one-shotted by ml-intern. someone created entire PhD dissertation chapter on context-aware agentic cyber defense drafted with 16 research subagents. and someone used it to crack an Paul Jankura kernel optimization take-home. (we don't know how to feel about this one 👀 ) just getting started →show more

We are entering an extremely exciting era for open-weight... models. Kimi K2.6 now feels like a top agentic model. I took it for a spin via Fireworks AI fast inference APIs. Kimi K2.6 has impressive agentic capabilities, design skills, and the ability to synthesize large amounts of information. I built a little Skill that produces survey papers on any AI research topic you want. (see example in the clip) You can use the skill to tell your agent to generate a survey on whatever topic and watch it go to work. The artifact was fully generated by Kimi.ai's Kimi K2.6. It's cheap and fast. Next step for me is to explore ways to continue integrating the capabilities of these models on use cases like automating my LLM knowledge bases and augmenting my agent memory capabilities. Stay tuned for more.show more

BREAKING: Anthropic just dropped Opus 4.8—and it is a... MONSTER We've been testing for about a week Every 🪨 and our verdict is they could've just called it Opus 5, it's that good. Here's our vibe check: - Beats GPT-5.5 on Senior Engineer bench. On our toughest benchmark Opus 4.8 scores a 63—a hair higher than GPT-5.5's score of 62, and a full 30 points higher than Opus 4.7. It tackled a ground-up rewrite of a production codebase, and actually built something that works. HOWEVER: Coding performance varied a lot at different reasoning levels. We recommend using it on xhigh for best results. - Incredibly good writer. Opus 4.8 scored a 79.6 on our writing benchmark—measuring models on real-world writing tasks we do all of the time like essay writing, promo email writing, and more. It beats GPT-5.5 by 6 points. It produces well-written prose with fewer "AI-isms". It's also very good at writing in your voice given the right context. HOWEVER: Writing performance also varied with reasoning levels. Medium reasoning had higher incidence of AI-isms—we found best results with high. - Beast at knowledge work. Opus 4.8 is very good at general knowledge work tasks like report creation, research and more. It produced the best PowerPoint one-shot we've ever seen on our deck generation benchmark. - Emotionally intelligent, willing to question the frame. I've also found it to be quite good at talking through psychological or interpersonal issues. It has a high EQ, and it's also good at not glazing and helping to expand your perspective. Its thought process feels extremely rich and dynamic. THE BAD: These days a model is only as good as its harness, and Codex is still a far superior harness to the Claude Desktop app. This has kept me using Codex + GPT-5.5 as my daily driver, but I am flipping back and forth a lot more between Codex and Claude. Anthropic is back baby! Read the rest on Every 🪨:show more

Fine-tune DeepSeek-OCR on your own language! (100% local) DeepSeek-OCR... is a 3B-parameter vision model that achieves 97% precision while using 10× fewer vision tokens than text-based LLMs. It handles tables, papers, and handwriting without killing your GPU or budget. Why it matters: Most vision models treat documents as massive sequences of tokens, making long-context processing expensive and slow. DeepSeek-OCR uses context optical compression to convert 2D layouts into vision tokens, enabling efficient processing of complex documents. The best part? You can easily fine-tune it for your specific use case on a single GPU. I used Unsloth to run this experiment on Persian text and saw an 88.26% improvement in character error rate. ↳ Base model: 149% character error rate (CER) ↳ Fine-tuned model: 60% CER (57% more accurate) ↳ Training time: 60 steps on a single GPU Persian was just the test case. You can swap in your own dataset for any language, document type, or specific domain you're working with. I've shared the complete guide in the next tweet - all the code, notebooks, and environment setup ready to run with a single click. Everything is 100% open-source!show more

THIS GUY GOT SICK OF UK TRAIN STATIONS HIDING... THE PLATFORM NUMBER UNTIL THE LAST SECOND, SO HE VIBE CODED AN APP THAT PREDICTS IT BEFORE THEY REVEAL IT if youve ever stood in a crowded station like london euston, you know the pain the platform stays hidden until 15 minutes before boarding, then it flashes up and hundreds of people sprint for the same gate so he built something that fixes this: > a clean departures and arrivals app pulling live data straight from the network rail apis > live train tracking so you can watch where your train actually is on the route and whether its running on time > the main feature is a predictive engine that guesses your platform before the station reveals it, and its right about 75% of the time heres how it predicts: every time a train finishes its journey, it logs which platform it actually pulled into and compares that to the platform network rail originally advertised it saves every one of those comparisons, and over time it builds up enough history to predict the platform with a real confidence score he built the whole thing with claude, mostly opus with some fableshow more

I just built a Meta Ads diagnostic in Claude... Code that tells you WHY your account broke, not just what changed 🤯 It spins up a team of agents that each investigate a different reason performance dropped, then argue against each other to kill the wrong answer before it ever reaches you. All inside Claude Code. Perfect for DTC brands and agencies who panic-kill creative the second CPA spikes. If you've watched ROAS fall off a cliff and opened Ads Manager with ten tabs going, you already know what happens next. Your gut says "creative fatigue." You kill your best-performing ad. A week later performance is still broken, because that was never the problem. Guessing wrong is the most expensive move in paid social. This workflow ends the guessing: → One agent investigates each competing theory — creative fatigue, budget and delivery changes, traffic quality, offer and seasonality → Each one is blind to the others, reasoning only from its own slice of the data so they can't bias each other → A refuter agent then attacks every surviving theory and tries to kill it → A theory only stands if the data can't disprove it → You get a ranked diagnosis: the real cause, the evidence for and against it, and the one move to make this week No anchoring on the first obvious answer. No killing winning creative on a hunch. No "here's what happened" reports that never tell you why. What you get: → Every theory tested in parallel instead of one biased guess → An adversarial pass that kills the wrong answer before you act on it → A ranked diagnosis with confidence levels and evidence both ways → A reusable workflow you drop next month's export into and re-run Built 100% in Claude Code with the new dynamic workflows. The first account I ran it on looked like textbook creative fatigue. The workflow disagreed, and traced the real cause to a budget change that had doubled spend and flooded delivery with junk traffic. I put together a full playbook with the exact workflow, the prompt, and how to run it on your own account. Want it for free? > Like this post > Comment "META" And I'll send it over (must be following so I can DM)show more

You don't understand... Higgsfield MCP + Claude just automated... AI film making. Every single step you used to grind through to make an AI movie, you can now do 10x faster. Drop the script into Claude Opus 4.8 and say: "Here's my script. Break it into a full shotlist. Shot number, scene, shot type, camera move and the action in each frame." Now the whole film is mapped, shot by shot. - Pull your assets. Ask Claude: "From this shotlist, list every character, every location and every prop across the whole film." That's your build list. The stuff you would need to generate and give as references in next steps. - Build the character sheets. Higgsfield MCP is connected, so Claude has hands now to do stuff directly. It generates the images itself. Have the full body, back view and close up in the character sheet. One per character. Each sheet becomes the locked reference for that face. Same move for locations, generate the empty plate for each one before anyone steps into it. - Generate the frames. Feed Claude the references plus the shot and have it write and fire the Seedance 2.0 prompt. "Using the lead's character sheet and the alley plate, generate shot 4 in Seedance 2.0. Low angle, slow push-in, rain." Claude builds the prompt, calls Seedance 2.0 and the frame lands back in chat. Use a Seedance 2.0 skill to teach Claude how to prompt it properly. Now, there are 3 ways to make the shots. Pick one per scene. - Pure prompting. Fastest one. You describe the action in words and let Seedance interpret it. For consistency across a sequence, feed it a frame from the previous shot so the look carries. - Storyboarding. You hand it a panel and it matches that composition exactly. Way more control over how the shot is framed. The tradeoff is that it can introduce more cuts than you actually want. - Path Control System This is the latest technique Seedance 2.0 technique. Generate a still base plate of the scene. Draw a red line across it to mark the exact path of the movement, then describe what's happening. Seedance follows that line for the action. Also ask Claude to remove the red line when animating. This is the one for anything where motion has to land precisely. The output reads like real live action. - Lastly, generate every clip you need, then cut them together. Get it to Capcut for editing and audio design. And that's it. The pipeline that used to need a full crew and a studio can now run from one Claude chat. 2026 is gonna be wildshow more

Building a personal knowledge base for my agents is... increasingly where I spend my time these days. Like Andrej Karpathy, I also use Obsidian for my MD vaults. What's different in my approach is that I curate research papers on a daily basis and have actually tuned a Skill for months to find high-signal, relevant papers. I was reviewing and curating papers manually for some time, but now it's all automated as it has gotten so good at capturing what I consider the best of the best. There are so many papers these days, so this is a big deal. You all get to benefit from that with the papers I feature in my timeline and on DAIR.AI. The papers are indexed using tobi lutke qmd cli tool (all of it in markdown files along with useful metadata). So good for semantic search and surfacing insights, unlike anything out there. I am a visual person, so I then started to experiment with how to leverage this personal knowledge base of research papers inside my new interactive artifact generator (mcp tools inside my agent orchestrator system). The result is what you see in the clip. 100s of papers with all sorts of insights visualized. I keep track of research papers daily, so believe me when I tell you that this system is absolutely insane at surfacing insights. This is the result of months of tinkering on how to index research and leverage agent automations for wikification and robust documentation. But this is just the beginning. The visual artifact (which is interactive too) can be changed dynamically as I please. I can prompt my agent to throw any data at it. I can add different views to the data. Different interactions. I feel like this is the most personalized research system I have ever built and used, and it's not even close. The knowledge that the agents are able to surface from this basic setup is already extremely useful as I experiment with new agentic engineering concepts. I feel like this knowledge layer and the higher-level ones I am working on will allow me to maximize other automation tools like autoresearch. The research is only as good as the research questions. And the research questions are only as good as the insights the agents have access to. Where I am spending time now is on how to make this more actionable. I am obsessed about the search problem here. The automations, autoresearch, ralph research loop (I built one months ago) are easier to build but are only as good as what you feed them. Work in progress. More updates soon. Back to building.show more

As always everyone is blind staring at the progress... of LLMs for coding and chat But meanwhile the new SOTA video model Seedance 2.5 has been slowly rolling out and it's really quite exceptional It's made by ByteDance (TikTok) who of course have lots of training data With just a few reference pics, it can get quite close to how you look IRL and you can do quite professional video shots with just a prompt I'd say it's the first video model that's now at the level of image models with the level of character likeness, cracking that in image models also took about 3 years (2022-2025) Generating 15 seconds takes about 4 minutes I put it live now on Photo AI, you can use it under [ Make video ] from the sidebar with just a prompt and your model selected! So you don't need to take an AI photo first and then turn that into a video! Saves lots of time :D It's more expensive than but I kept the credits the same (30 for 1 video) It also works inside the new video editor and you can make changes in your video with [ Magic edit ] in both the main app and the video editor Also a message for my server guy Daniel Lockyer (it can do voice too and you can even submit a voice sample of yourself, but I didn't here)show more

What if crypto research was as easy as chatting... with ChatGPT, but powered by real market data👑 Introducing CMC AI, a powerful new tool from CoinMarketCap that combines the speed of AI with the depth of live crypto data. It delivers fast, data-backed answers to your questions: ✅ Want to know why Bitcoin's price is rising? ✅ Curious about the latest news on your favorite cryptocurrency? ✅ Need sentiment analysis? It pulls real-time data and explains it in seconds. But it goes far beyond basic Q&A. In the future, you’ll be able to ask anything! For example, you could ask it to: – Discover undervalued tokens based on volume, MC, and sentiment. – Compare Layer 1s or L2s by adoption, speed, and dev activity. – Detect rug-pull risk via wallet distribution and tokenomics red flags. – Break down your portfolio by risk, correlation, and potential return. – Explore new use cases in DeFi, AI, RWA and DePIN And much more! 🔗Try it here: CMC AI changes how you learn, think, and act in Web3🧠show more

What is the best video editing agent for short... form social? Does it actually work? We watched professional video editors, step by step, as they built short-form social reels in Adobe Premiere Pro. Today we're open-sourcing this preview dataset on Hugging Face, to make AI agents better at editing videos. The data set is 234 annotated steps across 4 computer-use trajectories. Editors narrated their reasoning aloud as they worked, so every step pairs a screenshot with the expert's own thought, a structured action, and executable grounding: >a Premiere MCP tool call, keyboard shortcut, menu path, or coordinate click. >The format follows the AgentNet trajectory schema, extended with a Premiere action taxonomy and multi-path execution. ***That makes it directly usable for computer-use agent SFT, reasoning mid-training, tool-use and function calling, and benchmarking agents against a human expert baseline. Enjoy!show more

90% of "AI developers" just download pre packaged GGUF... files from Hugging Face, hit run, and call it a day. The top 10% know how to pull the raw safetensors, run the math, and quantize massive models into Q4_K_M themselves. If you think llama.cpp can only execute models, you’re missing the best part of the open source ecosystem. It’s a high performance optimization suite. Manually stripping 69% of the VRAM footprint off a brand new model architecture is where real infrastructure value is made. If you want to actually master local inference and deploy models like Google’s massive Gemma 4 12B it on consumer NVIDIA hardware using llama.cpp, you need to learn this pipeline. Let's build it. I just took the raw 22.7 GB Gemma 4 baseline and manually compressed it down to a 7.02 GB Q4_K_M GGUF artifact using llama.cpp. That is a 69% reduction in footprint. No quality loss. No VRAM bottlenecks. Just native, hardware accelerated C++ inference running a full 2,50,000 token context window on a dual NVIDIA Tesla T4 setup. Stop melting your VRAM on unoptimized weights and stop relying on other people's pipelines. Own your stack. I mapped this entire architecture from dynamic binary fetching to raw quantization and real time GPU streaming into a single, bulletproof notebook. Notebook link is in the comments below. Bookmark this blueprint for your next deployment and tell me which quantization works best for your workflow and model.show more

If you think this is just another silly demo... made with AI, read this post. You might change your mind, because this demo is about MATH. What you see on the screen is not a render from Blender (obviously, it’s not that good). It’s a three.js app built with Toolcraft. Available on the web and rendered in real time(link in the comments). But Blender still has a lot to do with it. Blender has Geometry Nodes - a powerful node-based system for creating and manipulating procedural geometry. In other words, it’s math. And math is a universal language. And who do you think is pretty good at math? >>> AI. Now you can download or buy Blender files from marketplaces, and when they contain Geometry Nodes for procedural animations, objects, surfaces, or effects, you can transfer that logic to the web. Make it real-time, make it interactive. Materials are a separate story, of course. They can still suck unless you use the right tricks: PBR, HDRIs, material blending, displacement, and faked surface relief. So why is Blender important here? Blender is open source, and many tools around it are open source too. An AI trained on their code. That means it can translate the math from one environment to another quite accurately. If you’ve been struggling to reproduce some idea with AI that you had in your head or seen in some references, and it has something to do with Geometry Nodes, and you can find that idea or a close one in the Blender ecosystem - it means you can transfer it to the web. Thank me later.show more

Here's what The Browser Company's AI eng & ML... teams are working on for Dia right now: (This is a pitch to come work for us; info at end) 🤖 COMPUTER USE – we've built our own bespoke APIs on top of Chromium to optimize latency, accuracy, and cost of computer-using agents. Demo attached. Big breakthroughs here in recent weeks. 🛡️ ON-DEVICE MODELS – we've built our own custom infra to run everything from encoder-only models to full LLMs on device. It's cross-platform, supports LoRa adapters, and optimized for the GPU. This system preserves privacy and enables fast inference times. 🧠 MEMORY – with your permission, Dia automatically tailors your AI experiences to you, personally, based on the tabs you open while browsing normally every day. We're also bringing vertical memory to specific features. ♻️ DATA FLYWHEELS – our Fall/Winter P0 is to double-down on training custom models based on implicit signals from daily use of Dia. Dia should get smarter and more useful the more people use it. Whether via RL, auto-generated prompts, or otherwise. If this work sounds interesting to you please visit our jobs page or email [email protected]. Hiring nearly every related role -- from ML engineers to people prototyping with AI and context/prompt writers -- everyone encouraged to apply!!show more