Xingchen Intelligence Releases Lumo-2
机器之心 · wechat · 2026-07-17
Xingchen Intelligence unveiled its second-generation embodied foundation model Lumo-2 and agent AgentPhilia on the eve of WAIC, releasing 20+ real robot videos showing dual-robot collaboration, egg frying, weighing, catching, coffee making, cocktail mixing, tying bows, zipping, folding clothes, and other complex household tasks.
The article emphasizes this is not a single demo but a full-stack approach integrating "AI model – embodied OS – cable-driven body": Lumo-2 predicts future states in a latent world before generating actions; through three-stage cross-modal alignment and chunked autoregressive decoding, end-to-end inference is 2.71x faster than standard autoregressive without sacrificing accuracy. AgentPhilia handles long-term memory, task scheduling, and multi-robot coordination, providing continuous service for home and enterprise scenarios.
Related event: Astribot launches Lumo-2 with real-robot demos(10 posts)→
More from Embodied
- RTX Spark laptops impress analyst: 30B local model runs quiet with great battery life — Scobleizer · 2026-10-27
- Re-uploading a 239GB robot dataset moved only 2.6MB — inside a petabyte-scale data pipeline — mervenoyann · 2026-10-27
- ROSCon UK photo tour: BBC's whale robot, da Vinci surgical robot and the OG PR2 — kamathsblog · 2026-10-27
- Teardown of a humanoid robot Z-axis linear actuator, with Zmotor Q&A — MarwaEldiwiny · 2026-10-27
- GPT-6 Astra Beats Gemini 3.8 Flash in Paired Robot Test, Yet Neither Completes the Task — qinzytech · 2026-10-27
- Teardown of Zmotor LA500 humanoid linear actuator: 324 g, 500 N peak thrust — MarwaEldiwiny · 2026-10-27