LingBot-VA SFT checkpoints trained with LoongForge, one per benchmark.
AI & ML interests
None defined yet.
Recent Activity
View all activity
Papers
Vision-Language-Delta-Action: Learning Compact Deltas as Actionable Transition Interfaces for Robotic Manipulation
TAOT: Topology-Aware Optimal Transport for Dynamic Expert Replica Placement in MoE Training
Organization Card
LoongForge is an open-source training framework developed by the Baidu AI Cloud Baige team, built to deliver faster training for mainstream LLMs, VLMs, diffusion, and embodied models, thereby significantly reducing costs.
models 8
LoongForge/LoongForge-LingBot-VA-RoboTwin
Robotics • 5B • Updated • 16
LoongForge/LoongForge-LingBot-VA-LIBERO
Robotics • 5B • Updated • 17
LoongForge/LoongForge-GR00T-N1.7-SimplerEnv-Fractal
Robotics • Updated • 44
LoongForge/LoongForge-GR00T-N1.7-SimplerEnv-Bridge
Robotics • Updated • 49
LoongForge/LoongForge-GR00T-N1.7-LIBERO
Robotics • Updated
LoongForge/Loongforge-pi05-RoboTwin
Robotics • 4B • Updated • 25
LoongForge/Loongforge-pi05-LIBERO
Robotics • 4B • Updated • 27
LoongForge/LoongForge-VLDA-LIBERO
Robotics • 7B • Updated • 46 • 2
datasets 0
None public yet