YccHugAi/lingbot-vla-2-stackcube
0
LingBot-VLA 2.0 — Stack-the-cubes finetune
Finetune of LingBot-VLA 2.0 (6B, MoE action expert) on the mixed LGG100/Stack-the-cubes + LGG100/Stack-the-cubes-v2 datasets (Unitree G1 Dex1, dual-arm cube stacking).
Code & real-robot serving: https://github.com/YCC-DAVID/robotic
Checkpoints
Each global_step_*/ folder holds the deployable HF-format weights (model-0000x-of-00006.safetensors + tokenizer/config), loadable directly.
Training
- Base: LingBot-VLA 2.0 (fp32), 16-dim state/action (14 arm joints + 2 grippers)
- 2xA100-80GB, FSDP2, Muon optimizer,
L1_fmloss,bounds_99_woclipnorm, absolute actions - 20000 steps, lr 1e-4, with depth + DINO-video distillation and future-image prediction
- Final loss ~0.11 (VLA_Loss)
Only the deployable hf_ckpt weights are published (optimizer states omitted).
