Spaces:
Running
Running
| title: README | |
| emoji: 😻 | |
| colorFrom: yellow | |
| colorTo: green | |
| sdk: static | |
| pinned: false | |
| <div align="center"> | |
| <div style="display: flex; align-items: center; justify-content: center; gap: 18px;"><img src="https://chunte-hfba.static.hf.space/images/Huggy%20Collection%202026/computer%20vision%20Huggy.png" alt="ml-intern" width="120"><h1 style="margin: 0; padding: 0; line-height: 1;">ML-intern-lab</h1></div> | |
| <div style="margin: 0; padding: 0; line-height: 1.2;"><strong>Models, datasets and demos built by an intern that never sleeps.</strong></div> | |
| <div style="margin: 4px 0 0 0; padding: 0; line-height: 1;"><a href="https://huggingface.co/chat"><img src="https://img.shields.io/badge/HuggingChat-ML%20Intern-ff9d00" alt="HuggingChat ML Intern"></a></div> | |
| </div> | |
| --- | |
| ## What is this? | |
| Everything in this organization was researched, trained, evaluated and pushed by [**ML Intern**](https://github.com/huggingface/ml-intern), the open-source agent from Hugging Face that autonomously researches, writes and ships ML code using the Hugging Face ecosystem, with access to docs, papers, datasets and cloud compute. | |
| The workflow for each repo: | |
| 1. **Describe the idea** in a conversation. | |
| 2. **The intern researches.** It searches the Hub, GitHub, papers and the web for the right models, datasets and tools. | |
| 3. **Approve a budget.** Once approved, the system enforces it. The agent can't change its own budget. | |
| 4. **The intern ships.** It creates the repo, generates data, launches training on HF Jobs, tracks the run on a Trackio dashboard, pushes the weights, writes the report and deploys a demo Space. | |
| What you see here is the output of that loop, unedited. | |
| --- | |
| ## What's in here | |
| | Collection | Contents | | |
| |---|---| | |
| | **🧠 Language models** | SFT, LoRA and GRPO runs on small open LLMs (MiniCPM5-2B, Qwen3.5, SmolLM) | | |
| | **👁️ Vision-language models** | VLM fine-tunes for OCR, document understanding and visual QA | | |
| | **🎨 Image LoRAs** | Style and subject adapters on FLUX.2 Klein, Qwen-Image and SD3.5 | | |
| | **📊 Datasets** | Synthetic and curated data the intern generated for its own runs | | |
| | **🚀 Demos** | Gradio Spaces auto-deployed at the end of each run | | |
| Each model card links to the session trace, the Trackio dashboard and the approved budget for that run. | |
| --- | |
| ## House rules | |
| - **Every run has a trace.** The full session log is stored as a Claude Code JSONL dataset and can be opened in the HF Agent Trace Viewer. | |
| - **Budget-capped.** Nothing here cost more than the amount approved before the run started. | |
| - **Evals as measured.** Runs that didn't beat the baseline are kept and reported as-is. | |
| - **Apache 2.0** or whatever the base model's license is. | |
| --- | |
| ## About | |
| This is not an official Hugging Face or ml-intern team org. The goal is to find out how far an autonomous ML agent can go on small models and small budgets. | |
| Built on: | |
| <div align="center"> | |
| [**ml-intern**](https://github.com/huggingface/ml-intern) · [**HuggingChat**](https://huggingface.co/chat) · [**HF Jobs**](https://huggingface.co/docs/huggingface_hub/guides/jobs) · [**Trackio**](https://github.com/gradio-app/trackio) · [**TRL**](https://github.com/huggingface/trl) · [**Gradio**](https://gradio.app) | |
| </div> | |
| <div align="center"> | |
| <sub>Reedi, Bonamy, Di Cosmo, von Werra, Tunstall. <em>ml-intern: an agent that autonomously researches, writes, and ships good quality ML related code using the Hugging Face ecosystem</em>, 2026.</sub> | |
| </div> | |