Hugging Face launches ml-intern open-source AI agent

Introducing ml-intern, the agent that just automated the post-training team @huggingface 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 http://hf.co/spaces, 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 http://hf.co/papers, reads them fully, walks citation graphs, pulls datasets referenced in methodology sections and on http://hf.co/datasets - 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: https://github.com/huggingface/ml-intern/tree/main Web + mobile: https://huggingface.co/spaces/smolagents/ml-intern And the best part? We also provisioned 1k$ GPU resources and Anthropic credits for the quickest among you to use.

• CL

  clem 🤗@ClementDelangue·2hOriginal post

  We built HF for AI builders collaboration, fun to see it's increasingly becoming the place for agent collaboration! This morning I'm sending my ml-intern to participate in the @OpenAI Parameter Golf challenge which goal is to train the best language model that fits in a 16MB artifact and trains in under 10 minutes on 8xH100s, evaluated by compression on the FineWeb validation set. What's super cool is that all ml-intern agents collaborate on the same @huggingface buckets, datasets, leaderboard and community tabs so they can communicate with each other for example to look at what the previous ones did, how and why they failed and hopefully add/improve/fix things each time a new agent joins the challenge (instead of starting from scratch everytime). Let's go agent collaboration! 🦾🦾🦾

  

  View on
• LE

  Lewis Tunstall@_lewtun·1hQuote tweet

  gpt-5.5 is now available in the ml-intern! this means it gets access to the whole @huggingface infra: buckets, jobs, repos etc for doing ai research at scale giving it a spin now to see if i'm even closer to automating myself :) https://github.com/huggingface/ml-intern/pull/118

  RO

  roon@tszzl·1dView on

  there are early signs of 5.5 being a competent ai research partner. several researchers let 5.5 run variations of experiments overnight given only a high level algorithmic idea, wake up to find completed sweep dashboards and samples, never having touched code or a terminal at all

  View on