Researchers have introduced RynnBrain 1.1, a new family of embodied foundation models available in 2B, 9B, and 122B-A10B scales. This model family is designed for robots, supporting perception, spatial reasoning, localization, and planning with a focus on physical grounding and manipulation. The 122B-A10B version has demonstrated superior performance on several benchmarks, outperforming existing proprietary and open-source models, and real-world robot experiments show RynnBrain-initialized policies are more effective than those based on Qwen or other generalist VLAs. AI
IMPACT Introduces advanced embodied AI capabilities for robotics, potentially improving robot manipulation and task performance.
RANK_REASON The cluster describes a new research paper detailing an embodied foundation model. [lever_c_demoted from research: ic=1 ai=1.0]
Read on Hugging Face Daily Papers →
- Astribot S1
- Maritime Mobile Service Identity
- Qwen
- RefSpatial-Bench
- RynnBrain 1.0
- RynnBrain 1.1
- RynnBrain-VLA
- Tianji-Wuji
- Unitree G1
- VSI-Bench
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
