Apollo World
Baidu launched Apollo World, an AI system that simulates real-world driving environments to train robotaxis and can generate up to 100,000 km of simulated driving per day.
Apollo World simulates road environments and predicts how they evolve, producing real-time multimodal data streams such as camera and lidar observations. Self-driving systems can sense and navigate the simulated environments directly. Baidu says it is already used to train the Apollo Go robotaxi fleet that will operate globally.
- Date
- Tuesday 3 November 2026
- Lab
- Baidu
- Kind
- model
- Access
- research preview
Figures
| Measure | Value | Measured by |
|---|---|---|
| Simulated driving per day | up to 100,000 km Baidu says this equals about three years of real-world testing by a safety-driver-equipped autonomous vehicle | company |
Based on Reuters coverage only. Access is an internal training tool for Apollo Go, and no external availability or price was stated.
Sources
This record was checked and corrected against its sources on 3 November 2026. How we check