@HuggingPapers: PhysBrain 1.5: Open-source SOTA for Embodied AI A single 8B model that understands scenes, generates robot actions, and…
Summary
PhysBrain 1.5 is an open-source 8B parameter model for embodied AI that achieves state-of-the-art performance by understanding scenes, generating robot actions, and predicting future frames, scoring 72.5 on 28 benchmarks.
View Cached Full Text
Cached at: 09/15/26, 11:49 PM
PhysBrain 1.5: Open-source SOTA for Embodied AI
A single 8B model that understands scenes, generates robot actions, and predicts future frames — all as tokens. 72.5 on 28 benchmarks, beating all open models. https://t.co/iTK3ZXRTzW
Similar Articles
PhysBrain 1.5: From Vision-Language Models to Physical Foundation Models
PhysBrain 1.5 is a unified model that integrates physical environment understanding, action generation, and future state prediction via autoregressive training, achieving state-of-the-art open-source performance on 28 embodied benchmarks.
PhysBrain 1.0 Technical Report
PhysBrain 1.0 is a technical report presenting a method that uses human egocentric video to generate physical commonsense supervision for vision-language-action models, achieving state-of-the-art results on embodied control benchmarks including ERQA, PhysBench, SimplerEnv-WidowX, LIBERO, and RoboCasa.
tencent/Hy-Embodied-RxBrain-1.0 · Hugging Face
Tencent releases Hy-Embodied-RxBrain-1.0, a unified multimodal foundation model for embodied cognition that combines language reasoning with visual imagination for understanding, world state prediction, and subgoal planning.
RynnBrain 1.1: Towards More Capable and Generalizable Embodied Foundation Model
RynnBrain 1.1 is a family of embodied foundation models (2B, 9B, 122B-A10B) that improve perception, spatial reasoning, and manipulation, achieving state-of-the-art results on VSI-Bench, MMSI, and RefSpatial-Bench, and outperforming baselines in real-robot experiments.
this looks promising: stepfun-ai/Step-5-Preview-BF16 · Hugging Face
StepFun AI has released a preview version of their Step-5 model in BF16 format on Hugging Face, indicating a promising development in AI model accessibility and performance.