Model family

Faster-WAM

Participating organizations

A world action model for language-guided robot manipulation. It computes a representation of future observations once and reuses it during action denoising. The authors provide separate LIBERO and RoboTwin checkpoints.

Overview

Model type
Action policy
Architecture
WAM
Listed versions
Faster-WAM

Model history

November 2026

WAMotor inference acceleration for Faster-WAM

New capabilityOfficial announcement

The WAMotor authors published a speculative inference implementation for Faster-WAM. It reuses denoising velocities without training a separate draft model and was evaluated on LIBERO, RoboTwin 2.0 and LIBERO-Plus.

Previously reported

The study uses this model’s standard flow-matching inference as its baseline.

This announcement

The authors used an RTX 5090 and measured inference latency per action chunk and mean task success. Each row compares standard inference with WAMotor. LIBERO (40 tasks × 50 trials) — 275.85 → 90.10 ms (3.06×), 98.60% → 99.20% RoboTwin 2.0 (50 tasks × 100 trials in each of clean and randomized settings) — 282.09 → 98.70 ms (2.86×), 92.45% → 92.05% LIBERO-Plus (10,030 instances × 3 trials across seven types of shift) — 282.62 → 90.33 ms (3.13×), 74.52% → 75.05%

Faster-WAM caches future context in one pass and uses sparse video–action interactions during speculative denoising. The WAMotor implementation defaults to 10 denoising steps, BF16 and Shift-5. Speculation starts after one Euler update, uses a verification threshold of 0.01 calibrated to the median step size and caps the window at the denoising step count. Enable it with --speculative. The LIBERO-Plus launcher defaults to a 15% task sample, seed 42 and one trial per instance. The numbers above come from the full study protocol. Upstream lists step 021700 for LIBERO and LIBERO-Plus, and step 029355 for RoboTwin 2.0. The WAMotor paper gives no exact checkpoint hashes. The repository provides a 2-camera LIBERO task config (also used for LIBERO-Plus) and a 3-camera RoboTwin task config. Their mode is one_pass_future_cache.

View all important updates →

Papers & reports (1)

Faster-WAM: Efficient Inference-Time Future Conditioning for Robust World Action Models

Paper available

arXiv: 2608.04404DOI: 10.48550/arxiv.2608.04404
Authors & affiliations (8)

Author order: Original order

  1. Weiheng Zhao

    Huazhong University of Science and Technology

  2. Haoyi Jiang

    Huazhong University of Science and Technology

  3. Xin Shi

    D-Robotics

  4. Liu Liu

    Horizon Robotics

  5. Fan Huang

    Xiamen University

  6. Zhizhong Su

    Horizon Robotics

  7. Wei Sui

    D-Robotics

  8. Xinggang Wang

    Huazhong University of Science and Technology

First published
2026-08-05
Latest paper revision
v1 · 2026-08-05
Venue
Not provided

Covered model versions

Paper ↗PDF ↗

Official resources

Submit a source or correction