Ant Group's Embodied Video-Action Model Tops Benchmarks
omarsar0 · x · 2026-07-11
Ant Group's embodied AI company @robbyantbrain released a video-action foundation model designed for robot control.
The results provided in the post include:
- An average score of 93.6 on RoboTwin 2.0, surpassing π0.5 and previous VA models
- Only a 0.6 point difference between clean and randomized scenarios
- Requires only 10–15 demonstrations to adapt, and supports cross embodiment transfer
Links to the project page and paper are included, representing a clear research/engineering advancement.
Related event: LingBot-VA/VLA 2.0 Released: Native Embodied Foundation Model(24 posts)→
More from Companies & People
- Anthropic Insiders: Not Everyone at the Lab Believes in High p(doom) — anpaure · 2026-09-11
- Lumara AI Film Festival Comes to NYC Oct 26, Top AI Filmmakers to Compete — 0xAllen_ · 2026-09-11
- X drama: Anthropic researchers accused of spying on academic customers and racing them to results — basedjensen · 2026-09-11
- Investor argues Palantir-Nvidia partnership should slash Anthropic's IPO valuation — pdamodaran · 2026-09-11
- a16z podcast: why 2-3 person startups are absent from policy debates — a16z Podcast · 2026-09-11
- PyTorch Day Korea 2026 launches first offline conf, CFP closes Sept 13 — PyTorch · 2026-09-11