Xiaomi MiMo-Embodied-7B published
Xiaomi released MiMo-Embodied-7B, a 7B parameter vision-language model for embodied AI on Hugging Face under MIT license. It is based on the Qwen2.5-VL architecture, served via Hugging Face Transformers, and supports text-generation-inference and endpoints.
PUBLISHED2025-11-19
OBSERVED2026-08-11
AGE9mo
SOURCES1
- 7B parameter vision-language model for embodied AI (image-text-to-text pipeline)
- Based on Qwen2.5-VL architecture, MIT license
- Served via Hugging Face Transformers, TGI and endpoints-compatible
- Referenced arxiv paper: 2511.16518
COMMUNITY
No curated reactions recorded for this event. Facts and takes are kept in separate layers — community context is added by hand, never blended into the record above.