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 deployment of embodied AI models, including vision-language-action (VLA) models and world-action models (WAMs). Existing inference runtimes target request-response serving and fall short of 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. Based on architectural analysis of representative VLA models and WAMs, Embodied.cpp captures a shared execution path organized 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, deploying across heterogeneous devices, robots, and simulators via a single backend abstraction. Evaluations on two VLA models (HY-VLA and pi0.5) and a preliminary WAM benchmark built on LingBot-VA Transformer blocks report task success rates of 100.0% and 91.0%, and a reduction of block memory from 312.2 MiB to 88.1 MiB respectively.

Overview

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

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.

Embodied.cpp

Embodied.cpp is a portable C++ inference runtime for embodied models. Based on an architectural analysis of representative VLA models and WAMs, it 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
Latency-first fused inference
Extensible operator and I/O support
Deployment on heterogeneous devices, robots, and simulators through a single backend abstraction

Results

Evaluations were conducted on two VLA models (HY-VLA and pi0.5) and a preliminary WAM benchmark based on LingBot-VA Transformer blocks:

VLA deployments achieved task success rates of 100.0% (HY-VLA) and 91.0% (pi0.5)
The WAM benchmark reduced block memory from 312.2 MiB to 88.1 MiB

These results show that Embodied.cpp significantly improves deployment efficiency while maintaining high accuracy.

--- *Auto-collected on 2026-07-05.*