Skip to content
RobotWorld
ZH
← Back to Videos

Möbius: Unified Motion Understanding and Generation

Loading video
Loading video

Möbius unifies text-to-motion, motion-to-text, motion completion and instruction-based editing in one model; the repo currently exposes the paper, demos and pipeline while code remains under internal review.

Motion-Language ModelHuman Motion GenerationMotion UnderstandingText-to-MotionMotion EditingHumanML3DMöbiusMotion Generation
Category: research
Author: @BeyondTheBoxCN
Date: 2026-11-06T00:00:00
Duration: 159.633s
Source: https://x.com/BeyondTheBoxCN/status/2118609934870815001
Reference: https://github.com/JiazhiShu/mobius-code

Related Videos

ActionPlan: Future-Aware Streaming Motion Synthesis via Frame-Level Action Planning
0:44

ActionPlan: Future-Aware Streaming Motion Synthesis via Frame-Level Action Planning

@XianghuiXie · 599 views · 2026-03-21
Motion GenerationMotion Diffusion
MotionBricks Now Runs in WebGPU on a Phone
1:05

MotionBricks Now Runs in WebGPU on a Phone

@repalash · 122 views · 2026-10-31
Motion GenerationWebGPU
PKU Motion-Omni End-to-End Dialogue System
0:08

PKU Motion-Omni End-to-End Dialogue System

@aigclink · 722 views · 2026-09-07
Embodied AIDigital Human
PINOC T2M vs NVIDIA ARDY Motion Comparison
0:12

PINOC T2M vs NVIDIA ARDY Motion Comparison

@Viggle_PINOC · 498 views · 2026-07-27
VigglePINOC
RobotWorld Tests Agents on 84 Robot Tasks
0:45

RobotWorld Tests Agents on 84 Robot Tasks

@JiankaiSun · 32 views · 2026-11-06
RobotWorldMultimodal Agents
A-0 Trains Robot Policies on 16,000 Hours of Human Video
0:39

A-0 Trains Robot Policies on 16,000 Hours of Human Video

@yh_kris · 53 views · 2026-11-04
Human demonstrationsRobot Learning

As an Amazon Associate, we earn from qualifying purchases.