Robotics
LeRobot
Safetensors
cae_fastwam
libero
world-action-model
File size: 4,977 Bytes
8ecb02e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
7c8d07f
8ecb02e
787e3d4
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
8ecb02e
 
 
 
 
7c8d07f
 
8ecb02e
 
 
 
 
 
 
7c8d07f
8ecb02e
 
 
7c8d07f
 
 
 
 
 
 
8ecb02e
 
 
 
 
 
 
 
 
 
7c8d07f
8ecb02e
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
---
library_name: lerobot
license: apache-2.0
pipeline_tag: robotics
base_model: amd/CAE-FastWAM
datasets:
- yuanty/LIBERO-fastwam
tags:
- cae_fastwam
- lerobot
- libero
- robotics
- world-action-model
---

# CAE-FastWAM (LeRobot format)

[CAE-FastWAM](https://huggingface.co/amd/CAE-FastWAM) converted to the [LeRobot](https://github.com/huggingface/lerobot) checkpoint format. CAE-FastWAM is a [FastWAM](https://arxiv.org/abs/2603.16666) world action model with a Compact Action Expert: a small (32-wide) action expert denoises latent action tokens, and a LIBERO-suite-conditioned action decoder turns them into the action chunk, on top of FastWAM's Wan2.2-TI2V-5B video expert.

The weights are the same as [amd/CAE-FastWAM](https://huggingface.co/amd/CAE-FastWAM); this repo only changes the layout: `config.json` (including the per-suite min/max normalisation statistics, `task_suite_stats`), `model.safetensors`, and FastWAM's standard LeRobot pre/post-processors, which use the statistics of the selected suite. It also carries the frozen action-VAE encoder, so the policy can be fine-tuned with the original CAE-FastWAM objective.

## Installation

The `cae_fastwam` policy type is proposed for LeRobot in [huggingface/lerobot#5317](https://github.com/huggingface/lerobot/pull/5317) and is not merged yet. Until then, install LeRobot from the commit used for this checkpoint:

```bash
pip install "lerobot[fastwam,libero] @ git+https://github.com/hiteshK03/lerobot.git@e0606546945b93f0cfe87c98918904279dba94ad"
```

Once the PR is merged, install from upstream instead:

```bash
pip install "lerobot[fastwam,libero] @ git+https://github.com/huggingface/lerobot.git@main"
```

Inclusion in a LeRobot release is decided by the LeRobot maintainers.

The script used to convert the original checkpoint is in [`converter/`](https://huggingface.co/amd/CAE-FastWAM-lerobot/tree/main/converter) (see its [README](https://huggingface.co/amd/CAE-FastWAM-lerobot/blob/main/converter/README.md)).

## Inputs and outputs

| Feature | Type | Shape |
| --- | --- | --- |
| `observation.images.image` (agent view) | visual | (3, 224, 224), resized by the model as in FastWAM |
| `observation.images.image2` (wrist view) | visual | (3, 224, 224), resized by the model as in FastWAM |
| `observation.state` | state | (8,) |
| `action` | action | (7,) |

Select the LIBERO suite with `--policy.task_suite` (`libero_spatial`, `libero_object`, `libero_goal` or `libero_10`); evaluate one suite per run.

## Results

LIBERO with LeRobot's `lerobot-eval`: bfloat16, 10 denoising steps, 256 x 256 camera frames, seed 42, 50 episodes per task (10 tasks per suite), at most 400 steps per episode (700 for `libero_10`):

| Suite | Success rate | Episodes |
| --- | ---: | ---: |
| libero_spatial | 96.0% (480/500) | 500 |
| libero_object | 99.4% (497/500) | 500 |
| libero_goal | 94.4% (472/500) | 500 |
| libero_10 | 95.8% (479/500) | 500 |
| **Average** | **96.4% (1928/2000)** | 2000 |

This version processes observations with FastWAM's own path (in-model image resize, LeRobot's standard min/max normaliser). The previous version of this repo, with its own image resize step and normaliser, scored 96.7% (1934/2000) in an earlier run of this protocol. Run side by side on the same 2000 episodes, the previous version scored 96.3% (1926/2000) and this version 96.4%; no task differed significantly.

Reproduce (per suite; `--env.episode_length=700` for `libero_10`), on one AMD Instinct MI355X:

```bash
lerobot-eval --policy.path=amd/CAE-FastWAM-lerobot --policy.task_suite=libero_spatial \
  --env.type=libero --env.task=libero_spatial \
  --env.observation_height=256 --env.observation_width=256 --env.episode_length=400 \
  --eval.batch_size=5 --eval.n_episodes=50 --seed=42
```

LIBERO stack: the one pinned by LeRobot's LIBERO benchmark image (hf-libero 0.1.4, robosuite 1.4.0, MuJoCo 3.3.1), CPU physics, OSMesa rendering. The MuJoCo version matters: with MuJoCo 3.8.1 the `libero_spatial` task 5 scene settles differently, and the previous version of this repo scored 94.5% on average under it. Please report the MuJoCo version with LIBERO results.

For reference, the [amd/CAE-FastWAM](https://huggingface.co/amd/CAE-FastWAM) card reports 97.28 ± 0.20% with GEAK-optimized inference and a GPU-simulated LIBERO (mean of 11 runs of 2000 episodes).

## Fine-tuning

The checkpoint holds the frozen action-VAE encoder and decoder, so `lerobot-train --policy.path=amd/CAE-FastWAM-lerobot --policy.task_suite=<suite> --accelerator.mixed_precision=bf16 ...` fine-tunes the video and action experts with the original CAE-FastWAM objective, one suite per run.

## Provenance and license

Converted from [amd/CAE-FastWAM](https://huggingface.co/amd/CAE-FastWAM) (Apache-2.0, Advanced Micro Devices, Inc.), a fine-tune of [Wan2.2-TI2V-5B](https://huggingface.co/Wan-AI/Wan2.2-TI2V-5B) with the [FastWAM](https://github.com/yuantianyuan01/FastWAM) recipe on LIBERO. See `LICENSE` and `NOTICE`.