Sep 6, 2026 · Artificial Intelligence

VLAct: 16 GPUs, 20% Data Beats GR00T N1.6 in Cross-Embodiment Transfer

VLAct introduces a representation-centric continued pre-training framework for Vision-Language-Action models, achieving 92.5% on RoboTwin 2.0 and surpassing all World Action Models on RoboDojo using only 16 GPUs and open data; with 20% downstream data it outperforms GR00T N1.6 on unseen GR-1 robot.

RoboDojoRoboTwinVLA

0 likes · 10 min read

VLAct: 16 GPUs, 20% Data Beats GR00T N1.6 in Cross-Embodiment Transfer

Machine Heart

Aug 3, 2026 · Artificial Intelligence

Does VLA Action Prediction Need an LLM? TurboVLA Achieves 32 Hz with 0.2 B Params on RTX 4090

TurboVLA, a real‑time vision‑language‑action model from Huazhong University of Science and Technology and Huawei, bypasses the large language model bottleneck by directly fusing visual and language features, achieving 32 Hz online action prediction on a single RTX 4090 with only 0.2 B parameters and 0.9 GB VRAM, while maintaining high success rates across LIBERO, RoboTwin 2.0, and real‑robot tasks.

LIBEROLLMRTX 4090

0 likes · 11 min read

Does VLA Action Prediction Need an LLM? TurboVLA Achieves 32 Hz with 0.2 B Params on RTX 4090

RoboTwin

VLAct: 16 GPUs, 20% Data Beats GR00T N1.6 in Cross-Embodiment Transfer

Does VLA Action Prediction Need an LLM? TurboVLA Achieves 32 Hz with 0.2 B Params on RTX 4090

Does VLA Action Prediction Need an LLM? TurboVLA Achieves 32 Hz with 0.2 B Params on RTX 4090