Event
Xiaomi Robotics unveils Xiaomi-Robotics-0 VLA model
Key points
- • Xiaomi Robotics introduced Xiaomi-Robotics-0, an open-source vision-language-action (VLA) model with 4.7 billion parameters.
- • The model was trained on a large dataset of cross-embodiment robot trajectories and vision-language data.
- • It achieved state-of-the-art performance across three simulation benchmarks and demonstrated real-world capabilities in tasks like towel folding and Lego disassembly.
Company context
Consumer electronics company producing smartphones, smart home devices, and IoT hardware at large scale. Operates an integrated hardware and software ecosystem with strong manufacturing and supply chain capabilities.
Context
- Company
- Xiaomi
- Segment
- Consumer
- Event type
- Demonstration
- Geography
- Beijing · China