ACT-2 Preview — first robotics model unifying generalization with reliability

Loading video

Tony Zhao's ACT-2 Preview: a single fine-tuning example teaches a new generalizable behavior — zero-shot, real unseen homes, 99% success rate.

Twitter VLA ACT-2 人形具身智能泛化端到端家庭服务

Category: perception

Author: @tonyzzhao

Date: 2026-07-16T00:00:00

Duration: 145.451s

Source: https://x.com/tonyzzhao/status/2077806003308179802

Reference: https://tonyzzhao.com/

Astribot Lumo-2 — 20+ household tasks, one foundation model, real robot

Astribot Lumo-2 — 20+ household tasks, one foundation model, real robot

@XRoboHub · 6 views · 2026-07-16

TwitterVLAAstribot

Early access to NVIDIA Isaac GR00T N1.7 is here 🎉 — an open, commerci...

Early access to NVIDIA Isaac GR00T N1.7 is here 🎉 — an open, commerci...

@NVIDIARobotics · 8 views · 2026-04-17

GR00THumanoidIsaac

MiniCPM-Robot open-sourced — 1.5B general-purpose VLA model

MiniCPM-Robot open-sourced — 1.5B general-purpose VLA model

@OpenBMB · 6 views · 2026-07-19

TwitterVLAOpenBMB

LiteReality-Agent turns room scans into interactive 3D scenes

LiteReality-Agent turns room scans into interactive 3D scenes

@ZheningHuang · 5 views · 2026-08-07

场景重建3D重建LiteReality

Self-Perception Without Motion Capture Cage

Self-Perception Without Motion Capture Cage

@enzelee1010 · 5 views · 2026-07-31

自主感知self-perception动捕

Air-powered vine robots dress the wearer (KAIST + Stanford)

Air-powered vine robots dress the wearer (KAIST + Stanford)

@vertonbiz · 6 views · 2026-07-18

Twitter人形藤蔓机器人