← Back to Videos

Galaxea G0.5 VLA Joins LeRobot

Loading video

Galaxea G0.5 is a 2B VLA model that predicts a subtask and object boxes before actions, while LeRobot can annotate plans, memory and visual QA for longer multi-step robot tasks.

VLA Humanoid LeRobot Galaxea Qwen3.5 Visual-Language-Action

Category: humanoid

Author: @LeRobotHF

Date: 2026-11-03T00:00:00

Duration: 26.3s

Source: https://x.com/LeRobotHF/status/2117641904175169763

OpenHLM drives every humanoid joint with one VLA

OpenHLM drives every humanoid joint with one VLA

@TongZha22057330 · 46 views · 2026-11-01

OpenHLMWholeBodyVLA

Gemini 4 Argon Builds Its Own Training Sims to Teach a Humanoid Cornhole

Gemini 4 Argon Builds Its Own Training Sims to Teach a Humanoid Cornhole

@GoogleAI · 209 views · 2026-10-13

Natura-1: A 50B MoE Foundation Model for General Dexterous Manipulation

Natura-1: A 50B MoE Foundation Model for General Dexterous Manipulation

@naturewill · 298 views · 2026-10-13

Foundation ModelsDexterous Manipulation

MobileVISTA: Pose Generalization from Single-Pose Demos

MobileVISTA: Pose Generalization from Single-Pose Demos

@s_wistreich · 220 views · 2026-10-07

Mobile ManipulationData Augmentation

Open-RAIL: China Mobile's Real-Time Execution Layer Between VLA Models and Humanoids

Open-RAIL: China Mobile's Real-Time Execution Layer Between VLA Models and Humanoids

@CyberRobooo · 1.1k views · 2026-09-23

LimX TRON2 VLA Demo EP.5: Two-Arm Autonomous Duck Assembly

LimX TRON2 VLA Demo EP.5: Two-Arm Autonomous Duck Assembly

@YJH_GIGIYE · 646 views · 2026-09-22

As an Amazon Associate, we earn from qualifying purchases.