PulseAugur
EN
LIVE 19:05:36

Embodied AI inference engine APXInf open-sourced by Wuwenxiong, Tsinghua, SJTU

By PulseAugur Editorial · [2 sources] ·

Wuwenxiong, in collaboration with Tsinghua University and Shanghai Jiao Tong University, has open-sourced APXInf, an inference engine designed for embodied AI on edge devices. This engine aims to optimize the performance of embodied models on robotic platforms, significantly reducing end-to-end latency from approximately 278ms to under 26ms for the PI 0.5 FP8 model on Jetson Thor. APXInf supports various hardware platforms including NVIDIA Jetson Orin, Jetson Thor, and RTX 4090, with plans to expand support to domestic chips and other hardware. AI

IMPACT This open-source engine could accelerate the deployment of embodied AI on robots by optimizing inference for edge devices, reducing latency and improving real-time control capabilities.

RANK_REASON Open-source release of a specialized inference engine for embodied AI, developed in collaboration with academic institutions.

Read on Pandaily →

AI-generated summary · Google Gemini · from 2 sources. How we write summaries →

Embodied AI inference engine APXInf open-sourced by Wuwenxiong, Tsinghua, SJTU

How we ranked this

Signal score
0 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Research
Open-source release of a specialized inference engine for embodied AI, developed in collaboration with academic institutions.
Source corroboration
2 independent sources
Multiple independent publishers reporting the same story raises confidence that it's real and newsworthy.
Topics
model release, infra
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
56 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

Full methodology in our editorial standards.

COVERAGE [2]

  1. 量子位 (QbitAI) TIER_1 中文(ZH) · 量子位的朋友们 ·

    Wuwenxiong, Tsinghua University, and Shanghai Jiao Tong University jointly open-source embodied edge inference engine APXInf, with Pi 0.5 performance SOTA

    卡位具身智能规模化落地“最后一公里”！

  2. Pandaily TIER_1 English(EN) · [email protected] (Pandaily) ·

    Infinigence AI Open-Sources APXInf for Embodied Edge Inference on Jetson Thor

    Infinigence AI open-sources APXInf with Tsinghua and SJTU: PI 0.5 FP8 on Jetson Thor drops end-to-end latency from ~278 ms to ~26 ms (~38.46 Hz), with Orin/Thor/RTX 4090 support and a domestic-chip roadmap.

RELATED ENTITIES

RELATED TOPICS