MiMo-Embodied-7B
Open 7B vision-language model covering both embodied AI and autonomous driving, said to lead 17 embodied and 12 driving benchmarks.
Xiaomi says it is the first open VLM to integrate two domains. One covers task planning, affordance prediction and spatial understanding, and the other covers perception, prediction and driving planning. MIT licence.
- Date
- Wednesday, 19 November 2025
- Lab
- Xiaomi (MiMo)
- Kind
- open-weights
- Access
- open weights
Figures
| Measure | Value | Measured by |
|---|---|---|
| Embodied AI benchmarks | 17 company-claimed | company |
| Autonomous driving benchmarks | 12 company-claimed | company |
Date from HF repo creation; 'first' claim is Xiaomi's.
Sources
This record was checked against its sources on 6 October 2026. How we check