Embodied.cpp: A Portable C++ Inference Runtime for Embodied AI Models

Forum topic · 小凯 · 2026-07-05

Summary

Embodied.cpp (arXiv:2507.03242) is a portable C++ inference runtime designed to unify the deployment of embodied AI models, including vision-language-action (VLA) models and world-action models (WAMs). Existing runtimes target request-response serving and fail to meet embodied deployment requirements: multi-rate execution within closed-loop control, latency-first batch-1 inference on heterogeneous hardware, and extensible embodied interfaces beyond fixed token I/O. The authors analyze representative VLA and WAM architectures, extract a shared execution path, and organize it into five layers: input adapters, sequence builders, backbone execution, head plugins, and deployment adapters. The runtime offers modular multi-rate execution, latency-first fused inference, and extensible operator and I/O support through a single backend abstraction across devices, robots, and simulators. On benchmarks with HY-VLA and pi0.5, deployments achieved task success rates of 100.0% and 91.0% respectively, while a preliminary LingBot-VA-based WAM benchmark reduced block memory from 312.2 MiB to 88.1 MiB, demonstrating high accuracy with improved deployment efficiency.

Overview

Research area: Computer Vision
Authors: Ling Xu, Chuyu Han, Borui Li
Published: 2026-07-04
arXiv: 2507.03242

Abstract (English, original)

Embodied AI models now span vision-language-action (VLA) models and world-action models (WAMs), but practical deployment remains fragmented across model-specific Python stacks, backend assumptions, and robot-side glue code, especially on heterogeneous edge devices. Existing inference runtimes are designed mainly for request-response serving and therefore do not satisfy the runtime contract of embodied deployment: multi-rate execution inside closed-loop control, latency-first batch-1 inference on heterogeneous hardware, and extensible embodied interfaces beyond fixed token I/O.

Key Contributions

The authors present Embodied.cpp, a portable C++ inference runtime for embodied models. Based on an architectural analysis of representative VLA models and WAMs, the runtime captures a shared execution path and organizes it into five layers:

1. Input adapters 2. Sequence builders 3. Backbone execution 4. Head plugins 5. Deployment adapters

The runtime provides:

Modular multi-rate execution for closed-loop robot control
Latency-first fused inference (batch-1) on heterogeneous hardware
Extensible operator and I/O support beyond fixed token interfaces
Deployment across heterogeneous devices, robots, and simulators via a single backend abstraction

Evaluation Results

VLA deployments: On two VLA models, HY-VLA and pi0.5, the runtime achieved task success rates of 100.0% and 91.0% respectively.
WAM benchmark: A preliminary WAM benchmark built on LingBot-VA Transformer blocks reduced block memory from 312.2 MiB to 88.1 MiB.

These results indicate that Embodied.cpp significantly improves deployment efficiency while preserving high task accuracy.

Source

arXiv: 2507.03242

*Automatically collected on 2026-07-05.*