LingBot-VLA 2.0 Robot Control Model
LearnWithBishal · x · 2026-07-11
LingBot-VLA 2.0 is a foundation model built from scratch for robot control, rather than being adapted from a general-purpose video model.
- Key design: visual tokenizer, action prediction, full causal Transformer
- Training/Generalization: adapts from 10–15 demonstrations and can transfer across different robot embodiments
- Results: achieves 93.6% on the RoboTwin 2.0 benchmark
- Author's emphasis: the goal is for the model to learn control-oriented representations directly, rather than just optimizing for "good-looking video effects"
The post also includes links to the project page and paper, pointing to LingBot-VLA 2.0 / Robbyant Foundation Model.
Related event: LingBot-VA/VLA 2.0 Released: Native Embodied Foundation Model(24 posts)→
More from Embodied
- Skyfall AI's SKY-001 world model turns sketches into CAD builds in 3.9s, vs 199s for GPT-6 Astra — rohanpaul_ai · 2026-10-27
- Tesla FSD approved in Latvia, the 10th European country; Musk praises smaller nations for moving fastest — elonmusk · 2026-10-27
- WanGP v18.81 adds Tencent's Hy-Nomas world model, playable on 8GB VRAM — cocktailpeanut · 2026-10-27
- NVIDIA open-sources FLORA, a robotics data engine ~125x cheaper than GPT-6 Astra — nvidia · 2026-10-27
- Embodied AI startups need a different operational discipline, investor says — ___Mufasaa · 2026-10-27
- π0.5 study: raising demos from 10 to 438 lifts Place Ball in Bowl success from 7% to 83% — stingning · 2026-10-27