Wuwenxiong, in collaboration with Tsinghua University and Shanghai Jiao Tong University, has released APXInf, an open-source inference engine designed for embodied AI on edge devices. This engine aims to bridge the gap between model capabilities and practical, large-scale robot deployment by optimizing for low latency and high throughput on limited hardware. APXInf supports platforms like NVIDIA Jetson Thor and RTX 4090, demonstrating significant reductions in end-to-end inference time and achieving high inference frequencies suitable for real-time robotic control. AI
IMPACT Accelerates the deployment of embodied AI on robots by optimizing inference for edge devices, enabling more responsive and real-time control.
RANK_REASON Open-source release of an inference engine for embodied AI, developed in collaboration with universities. [lever_c_demoted from research: ic=1 ai=1.0]
- APXInf
- $\ ext{PI}_{0.5}$
- Jetson Thor
- Python
- RTX 4090
- Rust
- Shanghai Jiao Tong University
- Tsinghua University
- Wuwenxiong
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
