wtf is maestro?

Maestro is a local AI studio for creating videos, images, and audio. It runs entirely on your own machine, so nothing is sent to the cloud. The main attraction is Director mode, where you supply an audio track or a story idea and a built-in language model plans out an entire music video or short film. It writes the screenplay, plans each shot to match the beat or pacing, generates starting frames, and then produces the final clips. For music videos, Director mode listens to your audio, figures out the tempo and structure, and times scene cuts to the beat. It can also transcribe spoken parts and tell different speakers apart, so you can assign shots to specific voices. For short films, it generates scenes with named characters, dialogue, and visual continuity across cuts. You can let it run the whole process automatically, or step in and review and edit each stage yourself. Studio mode gives you direct manual control over every supported model. You can generate video, images, audio, speech, music, and sound effects. It supports combining multiple clips, blending one video into another, and extending long videos. A built-in browser lets you search and install community add-ons called LoRAs, which modify what a model produces. Maestro can also generate a short guide for each LoRA explaining what it does and how to use it. Performance is handled with an auto-tune system that detects your graphics card and memory on first launch, then picks the right settings so you do not have to guess. If a generation runs out of memory, Maestro suggests a fix with a one-click button. A built-in local language model comes pre-configured and downloads automatically the first time you use it, with options to connect external providers instead. The README also documents workspaces for organizing separate projects, mature content controls that are off by default, and a dashboard for reviewing past Director runs. The full README is longer than what was shown.