README.md

metadata

license: apache-2.0
tags:
  - robotics
  - lingbot-va
  - unitree-g1
  - world-model

g1_putawaytools_rndchnk_2000step — LingBot-VA G1 post-trained transformer

Fine-tuned transformer for LingBot-VA on Unitree G1 (Dex1), task XiaoweiLinXL/unitree_put_away_tools_lerobot_v2: "Put the battery on the shelf labeled 'battery' and put the screwdriver on the shelf labeled 'Philips'."

• Base: robbyant/lingbot-va-base
• Post-training: 48 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). Per-step randomized chunk_size ∈ {1,2,3,4} and window_size ∈ {4..64} so the deployed model handles any chunk/window setting at inference (the "rndchnk" in the repo name).
• 8 GPUs × grad_accum=4 = effective batch 32, optimizer step 2000 of a 5000-step schedule.
• 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 EmbodyX/g1_putawaytools_rndchnk_2000step       --local-dir g1_pat_2000_dl

mkdir -p g1_pat_2000
ln -sf $(realpath g1_pat_2000_dl/transformer)   g1_pat_2000/transformer
ln -sf $(realpath lingbot-va-base/vae)          g1_pat_2000/vae
ln -sf $(realpath lingbot-va-base/text_encoder) g1_pat_2000/text_encoder
ln -sf $(realpath lingbot-va-base/tokenizer)    g1_pat_2000/tokenizer

Serve with CONFIG_NAME=g1_putawaytools MODEL_PATH=g1_pat_2000. transformer/config.json has attn_mode: torch (inference-ready).