-
sergiopaniego/Qwen3.5-2B-SmolDataEnvs-SFT-500
Image-Text-to-Text • 2B • Updated • 25 • 1 -
sergiopaniego/Qwen3.5-2B-SmolDataEnvs-SFT-1k
Image-Text-to-Text • 2B • Updated • 19 -
sergiopaniego/Qwen3.5-2B-SmolDataEnvs-SFT-full
Image-Text-to-Text • 2B • Updated • 21 -
sergiopaniego/smoldataenvs-opencode-traces
Viewer • Updated • 3.8k • 36
Sergio Paniego PRO
sergiopaniego
AI & ML interests
None yet
Recent Activity
updated a dataset about 6 hours ago
huggingface-projects/Deep-RL-Course-Certification updated a dataset about 12 hours ago
agents-course/final-certificates updated a dataset about 12 hours ago
agents-course/course-certificates-of-excellenceOrganizations
Slides from my talks
-
Building AI in the open (UC3M)
🤗Watch a talk on building AI in the open
-
Train the agent slides
🤖20Watch a talk on modern agent training techniques
-
Train the agent slides (PyTorch Madrid)
🤖Watch a talk on modern agent training techniques
-
Multi-harness RL slides
🎛12Watch a tutorial on training coding agents with RL environments
An off-the-shelf coding agent, left untouched, trained through its own harness with TRL and OpenEnv.
Training Agents Class 2 (distillation): script, pi-mono-chat dataset, distilled Qwen3 checkpoints, and the 2 Trackio dashboards.
In-depth technical articles and research pieces published by Hugging Face
-
The Ultra-Scale Playbook
🌌4.14kThe ultimate guide to training LLM on large GPU Clusters
-
The Smol Training Playbook
📚3.4kThe secrets to building world-class LLMs
-
Evaluation Guidebook
📝375Explore LLM benchmark score trends over time
-
FineVision: Open Data is All You Need
📝236A new open-source dataset for training VLMs
Vision Spaces created by me
Spaces where I've collaborated or that I consider unique!
Experiment from the blog huggingface.co/blog/sergiopaniego/mopd-trl: Open-MOPD models and data, the 60-step checkpoint and its Trackio dashboard
-
BytedTsinghua-SIA/Open-MOPD-SmolLM3-3B-MixSFT
Text Generation • 3B • Updated • 1.14k • 1 -
BytedTsinghua-SIA/Open-MOPD-SmolLM3-3B-RL-Math
Text Generation • 3B • Updated • 370 • 1 -
BytedTsinghua-SIA/Open-MOPD-SmolLM3-3B-RL-Code
Text Generation • 3B • Updated • 351 • 1 -
BytedTsinghua-SIA/Open-MOPD-SmolLM3-3B-RL-IF
Text Generation • 3B • Updated • 295 • 1
Training Agents Class 3 (GRPO): the training scripts, the Trackio dashboard and the task data.
Training coding agents inside RL environments with TRL and OpenEnv. Two demos, one per way of connecting an agent to a trainer.
Blog: https://huggingface.co/blog/sergiopaniego/bringing-carla-to-openenv-trl/
Spaces to compare vision models — there’s no single best model, only the best one for your specific use case.
This collection includes all the models, datasets and Spaces mentioned in the blog Vision Language Models: 2025 Update
-
sergiopaniego/Qwen3.5-2B-SmolDataEnvs-SFT-500
Image-Text-to-Text • 2B • Updated • 25 • 1 -
sergiopaniego/Qwen3.5-2B-SmolDataEnvs-SFT-1k
Image-Text-to-Text • 2B • Updated • 19 -
sergiopaniego/Qwen3.5-2B-SmolDataEnvs-SFT-full
Image-Text-to-Text • 2B • Updated • 21 -
sergiopaniego/smoldataenvs-opencode-traces
Viewer • Updated • 3.8k • 36
Experiment from the blog huggingface.co/blog/sergiopaniego/mopd-trl: Open-MOPD models and data, the 60-step checkpoint and its Trackio dashboard
-
BytedTsinghua-SIA/Open-MOPD-SmolLM3-3B-MixSFT
Text Generation • 3B • Updated • 1.14k • 1 -
BytedTsinghua-SIA/Open-MOPD-SmolLM3-3B-RL-Math
Text Generation • 3B • Updated • 370 • 1 -
BytedTsinghua-SIA/Open-MOPD-SmolLM3-3B-RL-Code
Text Generation • 3B • Updated • 351 • 1 -
BytedTsinghua-SIA/Open-MOPD-SmolLM3-3B-RL-IF
Text Generation • 3B • Updated • 295 • 1
Slides from my talks
-
Building AI in the open (UC3M)
🤗Watch a talk on building AI in the open
-
Train the agent slides
🤖20Watch a talk on modern agent training techniques
-
Train the agent slides (PyTorch Madrid)
🤖Watch a talk on modern agent training techniques
-
Multi-harness RL slides
🎛12Watch a tutorial on training coding agents with RL environments
Training Agents Class 3 (GRPO): the training scripts, the Trackio dashboard and the task data.
An off-the-shelf coding agent, left untouched, trained through its own harness with TRL and OpenEnv.
Training coding agents inside RL environments with TRL and OpenEnv. Two demos, one per way of connecting an agent to a trainer.
Training Agents Class 2 (distillation): script, pi-mono-chat dataset, distilled Qwen3 checkpoints, and the 2 Trackio dashboards.
Blog: https://huggingface.co/blog/sergiopaniego/bringing-carla-to-openenv-trl/
In-depth technical articles and research pieces published by Hugging Face
-
The Ultra-Scale Playbook
🌌4.14kThe ultimate guide to training LLM on large GPU Clusters
-
The Smol Training Playbook
📚3.4kThe secrets to building world-class LLMs
-
Evaluation Guidebook
📝375Explore LLM benchmark score trends over time
-
FineVision: Open Data is All You Need
📝236A new open-source dataset for training VLMs
Vision Spaces created by me
Spaces to compare vision models — there’s no single best model, only the best one for your specific use case.
Spaces where I've collaborated or that I consider unique!
This collection includes all the models, datasets and Spaces mentioned in the blog Vision Language Models: 2025 Update