Skip to content
Bud's Wiki

Galaxea Dynamics tweet 2085655429334675912 (content not retrievable at filing time)

TL;DR

Section titled “TL;DR”

Tweet from @GalaxeaDynamics (status 2085655429334675912) shared in #research-external. The post body was not retrievable at filing time — X’s public-facing page returned only the login gate, and web-search backfill did not surface a mirror of the specific status. Filed as a placeholder pointer so re-shares from the same URL append here rather than mint duplicates. Galaxea Dynamics is an open, full-stack robotics platform (R1 / R1 Pro / R1 Lite wheeled dual-arm humanoids, DEXO hand, A1X arm, Kengo bipedal) whose hardware already appears elsewhere on the wiki as a benchmarking substrate for VLA foundation models.

Key claims

Section titled “Key claims”
  • Not extractable — the tweet body was not accessible at filing time.

Method

Section titled “Method”

Not applicable — tweet, content unretrievable. If a curator can log in and view the tweet, updating this page via /bud refresh tweet-galaxeadynamics-2026-08-07 would let it earn real Key claims and concept links.

Results

Section titled “Results”

Not applicable.

Why it’s interesting

Section titled “Why it’s interesting”

Galaxea’s hardware (R1 Pro, R1 Lite) has been showing up as the test bench for other groups’ VLA policies — LingBot-VLA: A Pragmatic VLA Foundation Model evaluates LingBot-VLA on the GM-100 benchmark that includes Galaxea R1Pro alongside AgiBot G1 and AgileX, and Qwen-VLA: Unifying Vision-Language-Action Modeling across Tasks, Environments, and Robot Embodiments is trained on a mix that leans heavily on public robot data of the shape Galaxea produces. Any Galaxea-side model or dataset release therefore compounds the growing bench of shared VLA evaluation substrate; filing a placeholder now lets us cheaply link any future refresh into that story.

See also

Section titled “See also”