A musician used the pretrained Anticipatory Music Transformer (Thickstun et al., music-medium-800k) to improvise live alongside a MIDI keyboard without any fine-tuning, steering the model's sampling instead of training it. The player tracked their key while performing so the model could generate a second piano part in real time, demonstrating that a pretrained symbolic music model can follow a live human player through sampling control alone.
I wanted to see how far you can get making a pretrained symbolic music model "listen" to a live player without training it. Turns out pretty far. Setup: I play a MIDI keyboard, the Anticipatory Music Transformer (Thickstun et al., music-medium-800k) plays a second piano. While I play I track my key,
EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.