Embodied.cpp: A Portable Inference Runtime of Embodied AI Models on Heterogeneous Robots

Hugging Face Daily Papers Papers
embodied-ai inference-runtime edge-devices vision-language-action world-action-models heterogeneous-computing portable-runtime

Summary

Embodied.cpp is a portable C++ inference runtime that enables efficient deployment of vision-language-action and world-action models across heterogeneous edge devices and robots through modular execution layers and optimized inference.

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. We present Embodied.cpp, a portable C++ inference runtime for embodied models. Based on an architectural analysis of representative VLA models and WAMs, Embodied.cpp captures a shared execution path and organizes it into five layers: input adapters, sequence builders, backbone execution, head plugins, and deployment adapters. The runtime provides modular multi-rate execution, latency-first fused inference, and extensible operator and I/O support, enabling deployment across heterogeneous devices, robots, and simulators through one backend abstraction. We evaluate Embodied.cpp on two VLA models, HY-VLA and pi0.5, and on a preliminary WAM benchmark using a LingBot-VA Transformer block. The VLA deployments achieve successful closed-loop execution with 100.0% and 91.0% task success rates, respectively. The WAM benchmark reduces block memory from 312.2 MiB to 88.1 MiB. These results show that Embodied.cpp improves deployment efficiency while preserving high accuracy across diverse embodied model architectures.
Original Article
View Cached Full Text

Cached at: 07/06/26, 06:36 AM

Paper page - Embodied.cpp: A Portable Inference Runtime of Embodied AI Models on Heterogeneous Robots

Source: https://huggingface.co/papers/2607.02501

Abstract

Embodied.cpp is a portable C++ runtime that enables efficient deployment of vision-language-action and world-action models across heterogeneous edge devices through modular execution layers and optimized inference.

Embodied AI models now span vision-language-action (VLA) models andworld-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. Existinginference runtimes are designed mainly for request-response serving and therefore do not satisfy the runtime contract of embodied deployment:multi-rate executioninsideclosed-loop control, latency-first batch-1 inference on heterogeneous hardware, and extensibleembodied interfacesbeyond fixed token I/O. We present Embodied.cpp, a portable C++inference runtimefor embodied models. Based on an architectural analysis of representative VLA models and WAMs, Embodied.cpp captures a shared execution path and organizes it into five layers: input adapters, sequence builders, backbone execution, head plugins, and deployment adapters. The runtime provides modularmulti-rate execution, latency-firstfused inference, and extensible operator and I/O support, enabling deployment across heterogeneous devices, robots, and simulators through one backend abstraction. We evaluate Embodied.cpp on two VLA models, HY-VLA and pi0.5, and on a preliminary WAM benchmark using a LingBot-VA Transformer block. The VLA deployments achieve successful closed-loop execution with 100.0% and 91.0% task success rates, respectively. The WAM benchmark reduces block memory from 312.2 MiB to 88.1 MiB. These results show that Embodied.cpp improves deployment efficiency while preserving high accuracy across diverse embodied model architectures.

View arXiv pageView PDFGitHub17Add to collection

Get this paper in your agent:

hf papers read 2607\.02501

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper1

#### SEU-PAISys/Embodied.cpp Robotics• 5B• Updatedabout 4 hours ago • 1

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2607.02501 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2607.02501 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

Embodied-R1.5: Evolving Physical Intelligence via Embodied Foundation Models

Hugging Face Daily Papers

Embodied-R1.5 is a unified embodied foundation model that achieves state-of-the-art performance on 16 out of 24 embodied vision-language benchmarks using multi-task balanced reinforcement learning. It introduces a Planner-Grounder-Corrector closed-loop framework for long-horizon tasks and is open-sourced to facilitate future research.

Capek 0.5: An Execution-Centric Vision-Language Model for Embodied Intelligence

arXiv cs.AI

This paper presents Capek 0.5, an embodied vision-language model built around an execution-centric capability taxonomy, using specialist training via reinforcement learning followed by weight-space merging and routed policy-space distillation. It introduces a new state verification benchmark and demonstrates strong performance across embodied and general capabilities at 2B and 35B-A3B scales.

tencent/HY-Embodied-0.5

Hugging Face Models Trending

Tencent releases HY-Embodied-0.5, a suite of foundation models designed for embodied AI agents featuring a Mixture-of-Transformers (MoT) architecture with efficient 2B and powerful 32B variants for real-world robot control and spatial-temporal reasoning.