AI & ML interests

https://github.com/huggingface/cookbook

Recent Activity

sergiopaniegoย 
posted an update about 22 hours ago
view post
Post
449
The list of hands-on notebooks (some beginner-friendly!) to get started with fine-tuning using TRL keeps growing!!

โ€ข SFT
โ€ข GRPO
โ€ข Tool calling & agents
โ€ข RL environments with OpenEnv
โ€ข LLMs and VLMs
โœจ Many run on FREE Colab, making it super easy to get started fast!

https://github.com/huggingface/trl/tree/main/examples/notebooks
sergiopaniegoย 
posted an update 4 days ago
sergiopaniegoย 
posted an update 5 days ago
sergiopaniegoย 
posted an update 11 days ago
sergiopaniegoย 
posted an update 12 days ago
view post
Post
1918
The Christmas holidays are here! ๐ŸŽ„
Thinking about learning something new in AI?

@huggingface offers 12 FREE courses covering all the relevant topics, for every level of experience. A great challenge for the holidays (and worth saving for later ๐Ÿ™„)

Letโ€™s explore them!

๐Ÿง  ๐—Ÿ๐—Ÿ๐—  ๐—–๐—ผ๐˜‚๐—ฟ๐˜€๐—ฒ: large language models with HF tools
https://huggingface.co/learn/llm-course

๐Ÿค– ๐—”๐—ด๐—ฒ๐—ป๐˜๐˜€ ๐—–๐—ผ๐˜‚๐—ฟ๐˜€๐—ฒ: build and deploy AI agents
https://huggingface.co/learn/agents-course

๐ŸŽจ ๐——๐—ถ๐—ณ๐—ณ๐˜‚๐˜€๐—ถ๐—ผ๐—ป ๐—–๐—ผ๐˜‚๐—ฟ๐˜€๐—ฒ: diffusion models with ๐Ÿค— Diffusers
https://huggingface.co/learn/diffusion-course

๐Ÿ”Š ๐—”๐˜‚๐—ฑ๐—ถ๐—ผ ๐—–๐—ผ๐˜‚๐—ฟ๐˜€๐—ฒ: transformers for audio tasks
https://huggingface.co/learn/audio-course

๐ŸŽฎ ๐——๐—ฒ๐—ฒ๐—ฝ ๐—ฅ๐—Ÿ ๐—–๐—ผ๐˜‚๐—ฟ๐˜€๐—ฒ: deep reinforcement learning
https://huggingface.co/learn/deep-rl-course

๐Ÿ‘๏ธ ๐—–๐—ผ๐—บ๐—บ๐˜‚๐—ป๐—ถ๐˜๐˜† ๐—–๐—ผ๐—บ๐—ฝ๐˜‚๐˜๐—ฒ๐—ฟ ๐—ฉ๐—ถ๐˜€๐—ถ๐—ผ๐—ป ๐—–๐—ผ๐˜‚๐—ฟ๐˜€๐—ฒ: modern computer vision with HF
https://huggingface.co/learn/computer-vision-course

๐Ÿฆพ ๐—ฅ๐—ผ๐—ฏ๐—ผ๐˜๐—ถ๐—ฐ๐˜€ ๐—–๐—ผ๐˜‚๐—ฟ๐˜€๐—ฒ (๐—Ÿ๐—ฒ๐—ฅ๐—ผ๐—ฏ๐—ผ๐˜): learning-based robotics
https://huggingface.co/learn/robotics-course

๐Ÿงฉ ๐— ๐—–๐—ฃ ๐—–๐—ผ๐˜‚๐—ฟ๐˜€๐—ฒ: Model Context Protocol explained
https://huggingface.co/learn/mcp-course

๐Ÿงช ๐—” ๐—ฆ๐—บ๐—ผ๐—น ๐—–๐—ผ๐˜‚๐—ฟ๐˜€๐—ฒ: post-training AI models
https://huggingface.co/learn/a-smol-course

๐Ÿ•น๏ธ ๐— ๐—Ÿ ๐—ณ๐—ผ๐—ฟ ๐—š๐—ฎ๐—บ๐—ฒ๐˜€: AI in game development
https://huggingface.co/learn/ml-for-games-course

๐ŸงŠ ๐— ๐—Ÿ ๐—ณ๐—ผ๐—ฟ ๐Ÿฏ๐——: machine learning for 3D data
https://huggingface.co/learn/ml-for-3d-course

๐Ÿ“˜ ๐—ข๐—ฝ๐—ฒ๐—ป-๐—ฆ๐—ผ๐˜‚๐—ฟ๐—ฐ๐—ฒ ๐—”๐—œ ๐—–๐—ผ๐—ผ๐—ธ๐—ฏ๐—ผ๐—ผ๐—ธ: practical AI notebooks
https://huggingface.co/learn/cookbook

All of them can be found here: https://huggingface.co/learn
sergiopaniegoย 
posted an update 16 days ago
view post
Post
1821
Google DeepMind releases FunctionGemma, a 240M model specialized in ๐Ÿ”ง tool calling, built for fine-tuning

TRL has day-0 support. To celebrate, weโ€™re sharing 2 new resources:

> Colab guide to fine-tune it for ๐ŸŒ browser control with BrowserGym OpenEnv
> Standalone training script

> Colab notebook: https://colab.research.google.com/github/huggingface/trl/blob/main/examples/notebooks/grpo_functiongemma_browsergym_openenv.ipynb
> Training script: https://github.com/huggingface/trl/blob/main/examples/scripts/openenv/browsergym_llm.py (command to run it inside the script)
> More notebooks in TRL: https://huggingface.co/docs/trl/example_overview#notebooks
sergiopaniegoย 
posted an update 19 days ago
sergiopaniegoย 
posted an update 22 days ago
view post
Post
2099
๐ŸŽ„ last talk of the year about open AI and HF today at Universidad Rey Juan Carlos for undergrad students

always a pleasure to be back at my alma mater

๐ŸŽ… slides: https://github.com/sergiopaniego/talks
  • 1 reply
ยท
sergiopaniegoย 
posted an update 23 days ago
view post
Post
1676
TRL now includes agent training support for GRPOโ€ผ๏ธ

Train ๐Ÿ•ต๏ธ agents with ๐Ÿ”ง tools, enabling interaction with external functions and APIs.

And of course, a new notebook and scripts to get you up to speed

๐Ÿ“˜ notebook tutorial: https://github.com/huggingface/trl/blob/main/examples/notebooks/grpo_agent.ipynb

๐Ÿ“‚ script examples: https://github.com/huggingface/trl/blob/main/examples/scripts/grpo_agent.py

๐Ÿ“ฆ TRL v0.26.0 release: https://github.com/huggingface/trl/releases/tag/v0.26.0
  • 2 replies
ยท
sergiopaniegoย 
posted an update 24 days ago
view post
Post
2840
ICYMI, you can fine-tune open LLMs using Claude Code

just tell it:
โ€œFine-tune Qwen3-0.6B on open-r1/codeforces-cotsโ€

and Claude submits a real training job on HF GPUs using TRL.

it handles everything:
> dataset validation
> GPU selection
> training + Trackio monitoring
> job submission + cost estimation
when itโ€™s done, your model is on the Hub, ready to use

read more about the process: https://huggingface.co/blog/hf-skills-training
  • 1 reply
ยท
sergiopaniegoย 
posted an update 24 days ago
view post
Post
2267
We just released TRL v0.26.0!

It comes packed with updates:
> Agent training with tools in GRPO
> New CISPO & SAPO losses + reasoning rewards
> vLLM quantization in colocate mode
> Dataset shuffling in SFT
> Lots of NEW examples
> Tons of fixes and documentation improvements

  • 3 replies
ยท
sergiopaniegoย 
posted an update 25 days ago
sergiopaniegoย 
posted an update 29 days ago
view post
Post
2861
Want to get started with fine-tuning but donโ€™t know where to begin? ๐Ÿค“โ˜๏ธ

Weโ€™re expanding our collection of beginner-friendly free Colab notebooks so you can learn and fine-tune models using TRL at no cost

๐Ÿ”ฌ Check out the full list of free notebooks: https://huggingface.co/docs/trl/main/en/example_overview#notebooks

๐Ÿ”ฌ If you want more advanced content, we also have a lot to cover in the community tutorials: https://huggingface.co/docs/trl/community_tutorials

And now the obvious question: what would you like us to add next?
sergiopaniegoย 
posted an update about 1 month ago
view post
Post
2397
NEW: @mistralai released a fantastic family of multimodal models, Ministral 3.

You can fine-tune them for free on Colab using TRL โšก๏ธ, supporting both SFT and GRPO

Link to the notebooks:
- SFT: https://colab.research.google.com/github/huggingface/trl/blob/main/examples/notebooks/sft_ministral3_vl.ipynb
- GRPO: https://colab.research.google.com/github/huggingface/trl/blob/main/examples/notebooks/grpo_ministral3_vl.ipynb
- TRL and more examples: https://huggingface.co/docs/trl/index
  • 2 replies
ยท
sergiopaniegoย 
posted an update about 1 month ago
sergiopaniegoย 
posted an update about 1 month ago
view post
Post
3124
want to use open models easily through an API?

Inference Providers might be exactly what youโ€™re looking for sooo hereโ€™s a complete beginner-friendly walkthrough ๐Ÿง

https://www.youtube.com/watch?v=oxwsizy1Spw
  • 2 replies
ยท