g1_fdmv2_openLidPotato_500step — LingBot-VA G1 post-trained transformer

Fine-tuned transformer for LingBot-VA on Unitree G1 (Dex1), task bobchenyx/g1_dex1_open_lid_add_potato_033: "Open the pot's lid and put the potato inside the pot."

• Base: robbyant/lingbot-va-base
• Post-training: 50 demos, single task, lr 1e-5, FDM v2 recipe — mutually-exclusive per-microstep regime (rank-synced coin fdm_prob=0.5: FDM video-only L_fdm Eq.13 lambda_fdm=1.0 OR standard IDM L_dyn+L_inv; one forward, one backward). Optimizer step 500 of a 2000-step run (training still in progress).
• This repo contains only transformer/ — vae/, text_encoder/, tokenizer/ are unchanged from robbyant/lingbot-va-base.

Assemble an eval-ready checkpoint

hf download robbyant/lingbot-va-base                   --local-dir lingbot-va-base
hf download armanakbari4/g1_fdmv2_openLidPotato_500step --local-dir g1_olp_500_dl

mkdir -p g1_olp_500
ln -sf $(realpath g1_olp_500_dl/transformer)  g1_olp_500/transformer
ln -sf $(realpath lingbot-va-base/vae)        g1_olp_500/vae
ln -sf $(realpath lingbot-va-base/text_encoder) g1_olp_500/text_encoder
ln -sf $(realpath lingbot-va-base/tokenizer)  g1_olp_500/tokenizer

Serve with CONFIG_NAME=g1_openlidpotato MODEL_PATH=g1_olp_500. transformer/config.json has attn_mode: torch (inference-ready).