An off-the-shelf coding agent, left untouched, trained through its own harness with TRL and OpenEnv.
Sergio Paniego PRO
sergiopaniego
AI & ML interests
None yet
Recent Activity
upvoted a collection about 1 hour ago
SmolDataEnv Multi-harness RL liked a Space about 1 hour ago
AdithyaSK/multi-harness-rl published a Space about 4 hours ago
sergiopaniego/tau2-envOrganizations
Training Agents Class 2 (distillation): script, pi-mono-chat dataset, distilled Qwen3 checkpoints, and the 2 Trackio dashboards.
-
sergiopaniego/pi-mono-youtube-livestream-2-scripts
Updated β’ 2 -
sergiopaniego/pi-mono-chat
Viewer β’ Updated β’ 886 β’ 41 -
sergiopaniego/qwen3-0.6b-pimono-logit-kd-lr1e5
Text Generation β’ 0.6B β’ Updated β’ 9 -
sergiopaniego/qwen3-0.6b-pimono-logit-kd-lr2e5
Text Generation β’ 0.6B β’ Updated β’ 8
In-depth technical articles and research pieces published by Hugging Face
-
The Ultra-Scale Playbook
π4.05kThe ultimate guide to training LLM on large GPU Clusters
-
The Smol Training Playbook
π3.31kThe secrets to building world-class LLMs
-
Evaluation Guidebook
π361Explore LLM benchmark scores over time
-
FineVision: Open Data is All You Need
π236A new open-source dataset for training VLMs
Vision Spaces created by me
Spaces where I've collaborated or that I consider unique!
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/
-
CARLA Environment Server
πControl a Carla driving simulation with custom actions
-
CARLA Environment Server
πControl a CARLA driving simulator with custom actions
-
Carla Grpo Trolley
πVisualize your programβs I/O activity in real time
-
sergiopaniego/Qwen3-0.6B-carla-trolley-escape
0.8B β’ Updated β’ 15
Spaces to compare vision models β thereβs 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
-
Qwen/Qwen2.5-Omni-7B
Any-to-Any β’ 11B β’ Updated β’ 332k β’ 1.95k -
Qwen2.5 Omni 7B Demo
π375Chat with text, audio, images, and video, get spoken replies
-
Qwen2.5-Omni Technical Report
Paper β’ 2503.20215 β’ Published β’ 174 -
openbmb/MiniCPM-o-2_6
Any-to-Any β’ 9B β’ Updated β’ 321k β’ 1.3k
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.
-
sergiopaniego/pi-mono-youtube-livestream-2-scripts
Updated β’ 2 -
sergiopaniego/pi-mono-chat
Viewer β’ Updated β’ 886 β’ 41 -
sergiopaniego/qwen3-0.6b-pimono-logit-kd-lr1e5
Text Generation β’ 0.6B β’ Updated β’ 9 -
sergiopaniego/qwen3-0.6b-pimono-logit-kd-lr2e5
Text Generation β’ 0.6B β’ Updated β’ 8
Blog: https://huggingface.co/blog/sergiopaniego/bringing-carla-to-openenv-trl/
-
CARLA Environment Server
πControl a Carla driving simulation with custom actions
-
CARLA Environment Server
πControl a CARLA driving simulator with custom actions
-
Carla Grpo Trolley
πVisualize your programβs I/O activity in real time
-
sergiopaniego/Qwen3-0.6B-carla-trolley-escape
0.8B β’ Updated β’ 15
In-depth technical articles and research pieces published by Hugging Face
-
The Ultra-Scale Playbook
π4.05kThe ultimate guide to training LLM on large GPU Clusters
-
The Smol Training Playbook
π3.31kThe secrets to building world-class LLMs
-
Evaluation Guidebook
π361Explore LLM benchmark scores over time
-
FineVision: Open Data is All You Need
π236A new open-source dataset for training VLMs
Vision Spaces created by me
Spaces to compare vision models β thereβs 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
-
Qwen/Qwen2.5-Omni-7B
Any-to-Any β’ 11B β’ Updated β’ 332k β’ 1.95k -
Qwen2.5 Omni 7B Demo
π375Chat with text, audio, images, and video, get spoken replies
-
Qwen2.5-Omni Technical Report
Paper β’ 2503.20215 β’ Published β’ 174 -
openbmb/MiniCPM-o-2_6
Any-to-Any β’ 9B β’ Updated β’ 321k β’ 1.3k