Sol plays Asterix from screenshots, holding each move until its next reply arrives

wzenus · x · 2026-10-17

Sol plays Asterix purely from screenshots, directly choosing movement commands; each direction stays held until its next reply arrives in real time. Through autoresearch it learns to prioritize escape over chasing food, finish a reachable pickup before switching targets, and pick a new direction after a pickup instead of repeating the held action. Another game environment from the Sooool "agent as real-time policy" project.

Related event: Developer Lets GPT-6 Sol Play Arcade Games in Real Time From Screenshots(2 posts)→

Original post →

More from coding & agent

coding & agent channel →