AgiBot WITA-Omni Ranks #1 on DailyOmni Benchmark

Loading video

AgiBot's general AI model WITA-Omni ranked #1 overall on the DailyOmni benchmark, topping all six subcategories including audio-visual understanding, temporal reasoning, and multimodal processing, beating Google Gemini 3.1.

多模态 Omni Model 具身智能 Twitter

Category: embodied

Author: @aarai666

Date: 2026-07-28T00:00:00

Duration: 103.4s

Source: https://x.com/aarai666/status/2082088536434745832

LightNav-0: Light Origins Unveils a Generalist Navigation Brain for Robots

LightNav-0: Light Origins Unveils a Generalist Navigation Brain for Robots

@LightOrigins_ · 28 views · 2026-08-26

机器人导航通用导航模型具身智能

Microsoft Fara 1.5 Computer-Use Models (4B/9B/27B)

Microsoft Fara 1.5 Computer-Use Models (4B/9B/27B)

@HuggingApps · 4 views · 2026-07-28

Computer UseGUI Agent多模态

Atom humanoid livestreams 3+ hours autonomously grasping and orienting packages

Atom humanoid livestreams 3+ hours autonomously grasping and orienting packages

@yuji_fujima · 21 views · 2026-08-31

Atomアトム人形机器人

Galaxea Unveils NEXO: 30-DoF Wheeled Dual-Arm Manipulator

Galaxea Unveils NEXO: 30-DoF Wheeled Dual-Arm Manipulator

@GalaxeaDynamics · 33 views · 2026-08-31

GalaxeaNEXO轮式双臂

Booster T2 Flagship Embodied Dev Platform Now Available

Booster T2 Flagship Embodied Dev Platform Now Available

@boosterobotics · 94 views · 2026-07-10

Booster加速进化Booster T2

Riemann-1.0 trains a World Action Model on 200k hours of human video

Riemann-1.0 trains a World Action Model on 200k hours of human video

@techniahqrobot · 74 views · 2026-08-29

World Action Model世界动作模型具身智能