Keygen Bench: Model changelog
85 ranked AI models write keygen tunes in FastTracker II, scored from the audio. Top: gpt-6-sol 73.7, gpt-6.1-sol 71.9, mimo-v2.6-pro 71.7.
Keygen Bench asks AI models to compose a keygen-style chiptune in FastTracker II, with no network and a Bash tool, three independent attempts each at its highest declared reasoning tier. An AI listener (Meta Audiobox Aesthetics) rates how enjoyable a trusted FT2 render of each tune is, and models are ranked by their best of three attempts.
Rankings | Tracker | Scoring | Changelog | Support