Xpeng Unveils GroundingPI Model for Robot Perception

Nov 4, 2026
Xpeng introduces GroundingPI, an open source embodied perception model with about 4 billion parameters. The company says it outscored OpenAI's GPT-6 Astra across 34 public perception benchmarks.

Xpeng unveils GroundingPI, an embodied perception foundation model with about 4 billion parameters, according to CnEVPost. The company says GroundingPI scored 73.68 across 34 public perception benchmarks, compared with 71.54 for OpenAI's GPT-6 Astra.

GroundingPI focuses on visual localization and spatial perception for robots. Its training uses perception data without robot joint action labels and covers 11 capability categories, including object detection, text recognition and robot spatial localization.

The model remains a scientific prototype. Xpeng says a simulation system using GroundingPI exceeded the execution success rate of comparison models trained with 75% of the demonstration data while using 50% of the data. The company has released the model weights and code as open source software.

We hope you enjoyed this article

Consider subscribing to one of our newsletters like Robotics Brief or Daily AI Brief.

Also, consider following us on social media:

AI & Robotics AI Brief AI Brief (X)
Free newsletter

Robotics Brief

Weekly coverage of AI-driven robotics advances in industrial automation, autonomous vehicles, and robotic systems.

Guide
Private humanoid makers by latest disclosed equity round

Private humanoid makers by latest disclosed equity round

Thirteen independent humanoid robot makers, ordered by the size of their latest disclosed round

Whitepaper

Tensordyne Napier: What If One Rack Could Do the Work of Nine?

Tensordyne

This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.

Read more

You may also like

Free, six days a week

Daily AI Brief: the AI news that matters, in your inbox.