Norman Di Palo (Google DeepMind): Gemini 4 Argon taught Apptronik’s Apollo humanoid cornhole from one human video
Norman Di Palo @normandipaloX
Why it matters
First-hand account by a Google DeepMind robotics researcher of the Argon cornhole demo, and the most viewed Argon launch post in the sweep list (~146k views).
Summary
Views ~146k, likes ~1.9k via api.fxtwitter.com on 2026-10-14 (quick run). Di Palo says that, given only a video of a person playing cornhole, Gemini 4 Argon took control of the Apollo humanoid, generated its own simulation environment, did system identification and trained throws on the real robot; after a few real shots it landed “many consecutive airmails”. At the end the robot explains what it did. Not opened on x.com.
Archived text
I’m still half in disbelief We asked Gemini Argon to teach the Apollo humanoid to play cornhole, and it zero-shotted it
All we gave it was a video of a person doing it. It then took control of the robot, generated its own sim env, performed some sysID, and trained launches on the real robot.
It took it a few real shots to improve and then land many consecutive airmails.
The process was so long and articulated that we asked the robot itself to explain what it did to us, as you can see at the end of the video. It even came up with the gestures
Great demo for a great model @googledeepmind @apptronik