Help & support.
Local AI music that stays local. Here's how to get help with it.
Email a human.
direct · no ticket queueEmail us about anything — a bug, a question about the model download, generation, training, or a feature request. Because Ohm Studio runs entirely on your device, please don't attach any private audio to your message; we can help without it. For non-private questions you can also open an issue on GitHub.
Email support ↵The short answers.
privacy · models · pricingNo. Generation and every file you import or export stay on your device; Mac training and stem separation stay on your Mac. There's no cloud LLM, account, or telemetry. The only model traffic is a consent-gated download from the repository disclosed before transfer.
Only for model setup. On iPad, the supported ACE-Step 1.5 assets come from a pinned Hugging Face revision after you review size, license, storage, integrity, and device fit. Mac uses its managed local catalogue. Once the required weights are verified locally, generation runs offline.
ACE-Step 1.5, an open music-generation model, running locally on your Apple Silicon GPU. Prompts, reference audio, and generated tracks are all processed and stored on your machine — nothing is sent to a hosted inference service.
Yes. Ohm Studio supports on-device LoRA training, so you can adapt the model to your own material without uploading a single sample. Training runs as a local background job on your Mac, and the resulting adapters stay in your local model directory.
Yes. Ohm Studio includes local stem separation built on the Ultimate Vocal Remover model family. The separation weights are an optional, on-demand download; once cached, splitting runs entirely on your Mac.
Yes. Tracks are generated on your own hardware from an open model, with no per-track licensing, no watermark, and no usage meter. What you generate is yours to use.
Ohm Studio is a one-time $19.99 App Store purchase for Mac and iPad. No subscription, credits, or per-generation fees.