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

# g1_fdmV2_allTasksLCM_500 — LingBot-VA G1 LCM-distilled transformer (step 500)

LCM video-only consistency distillation of the joint 5-task G1 alltasks
teacher (`armanakbari4/g1_fdmV2_allTasks_7500` — step 7500 of the
10000-step FDM-v2 post-training run on `JingwuLuo/all_tasks_lerobot`).
Target: 2-step video generation.

This repo ships the **target student** (EMA-frozen — the standard LCM eval
target). The online student exists in our training output but is not
uploaded; ask if you want it for comparison.

- Teacher: `armanakbari4/g1_fdmV2_allTasks_7500` (transformer/)
- Recipe (`distill_video_v2/config_g1_alltasks.py`):
  - distill_mode: `video` (video-only consistency loss)
  - `num_ddim_timesteps=2` (k=500 stride → target 2-step generation)
  - `lcm_skip_k=6`, EMA decay 0.995, huber loss (`huber_c=0.001`)
  - Teacher CFG range [2.0, 10.0]
  - lr=5e-6, grad_accum=8, batch=1, 4×H100
- Optimizer **step 500** of a 2000-step run (only 500/1000 were saved before
  the run was stopped at step ~1107; 1500/2000 were never reached).
- This repo contains **only the `transformer/`** (LCM-distilled, EMA target)
  — `vae/`, `text_encoder/`, `tokenizer/` are unchanged from
  `robbyant/lingbot-va-base`.

## Tasks covered (instruction strings used during teacher training)

| slug | instruction |
|---|---|
| `open_lid_add_potato` | Open the pot's lid and put the potato inside the pot. |
| `pick_red_bottle` | Pick up the red bottle |
| `pick_and_move_bottle` | Pick the pink object and put it on the cross mark. |
| `put_carrot_n_cup` | Pick up the carrot and put it inside the blue cup, then put the cup on the cross mark. |
| `put_cup_n_broccoli` | Pick the pink object and put it in the orange basket, then pick up the broccoli and put it inside the pink object. |

## Assemble an eval-ready checkpoint

```bash
hf download robbyant/lingbot-va-base                  --local-dir lingbot-va-base
hf download armanakbari4/g1_fdmV2_allTasksLCM_500       --local-dir alltaskslcm_500_dl

mkdir -p g1_alltasksLCM_500
ln -sf $(realpath alltaskslcm_500_dl/transformer) g1_alltasksLCM_500/transformer
ln -sf $(realpath lingbot-va-base/vae)              g1_alltasksLCM_500/vae
ln -sf $(realpath lingbot-va-base/text_encoder)     g1_alltasksLCM_500/text_encoder
ln -sf $(realpath lingbot-va-base/tokenizer)        g1_alltasksLCM_500/tokenizer
```

Serve with `CONFIG_NAME=g1_alltasks MODEL_PATH=g1_alltasksLCM_500` and set
`num_inference_steps=2` (the distillation target).
`transformer/config.json` has `attn_mode: torch` (inference-ready).