theDAW (Pinokio launcher)

One-click launcher for theDAW, the all-in-one AI music studio by GANTASMO. theDAW combines Stable Audio 3 and Magenta RealTime 2 generation, the Chimera multi-track fusion engine, Demucs stem separation, MIDI and notation tooling, DJ and VJ performance rigs, DAW project import (Ableton Live, Reaper, FL Studio, Audition, Bitwig, Resolume), VST3 and .gan plugin hosting, and a RAG-backed in-app assistant.

What the launcher does

• Install clones the repo into app/, pulls the Magenta sidecar submodule, installs FFmpeg through conda, resolves all Python dependencies with uv sync --group dev, and installs the frontend packages with npm install.
• Start launches the FastAPI backend on http://localhost:8600, then the Vite frontend on http://localhost:5173, and opens the app once the URL appears.
• Update pulls the launcher and the app repos, refreshes the submodule, and re-syncs Python and npm dependencies.
• Reset deletes app/.venv and app/frontend/node_modules so the next Install starts from clean dependencies.

Models are not downloaded at install time. theDAW runs in local-only mode by default and downloads a model the first time a generation needs it. The launcher points HF_HOME at the standard user Hugging Face cache (~/.cache/huggingface) rather than an isolated per-app cache, so checkpoints and the Hugging Face auth token already on the machine are reused. The Stable Audio 3 and t5gemma repos are gated, so a first-ever download needs a Hugging Face token (hf auth login, or the in-app download manager).

Platform behavior

The Python dependency set self-selects per platform through uv:

Platform	Torch build	Notes
Windows	CUDA 12.8 wheels + prebuilt flash-attention	Full feature set
Linux x86_64	CUDA 12.6 wheels	Full feature set; Magenta sidecar supported
macOS	Standard PyPI torch (CPU / MPS)	Small model recommended; flash-attention, Azure Kinect, and the Magenta sidecar are skipped automatically

The Small generation model runs on CPU, so machines without an NVIDIA GPU still generate audio.

Ports

The app fixes its own ports: the frontend proxies /api to localhost:8600 and Vite runs with strictPort on 5173. If Start fails immediately, close anything already using 5173 or 8600 (for example a copy launched through theDAW.bat). The VJ sidecar (port 5187) is spawned by the backend on first use and bootstraps its own npm packages.

The frontend's own dev script binds 0.0.0.0, so the web UI (and LAN features that talk to it, like the phone camera source) is reachable from other devices on the network. The backend binds to localhost under this launcher; features that contact the backend directly from another device (Quest streaming, XR control) need the theDAW.bat launch path, which binds the backend to 0.0.0.0.

Using the app

1. Click Install and wait for the dependency sync to finish.
2. Click Start. The backend comes up first, then the web UI; the Open App tab appears when the URL is ready.
3. Generate in MAKE, edit in EDIT, mix in MIX, perform in DJ / VJ / PERFORM. The in-app assistant and docs cover the rest.

API

The backend serves a full HTTP API on http://localhost:8600, with interactive documentation at http://localhost:8600/docs.

Generate audio

curl:

curl -X POST http://localhost:8600/api/generate \
  -F "prompt=warm analog synth arpeggio, 120 bpm" \
  -F "duration=30" \
  -F "steps=8"

Python:

import requests

r = requests.post(
    "http://localhost:8600/api/generate",
    data={"prompt": "warm analog synth arpeggio, 120 bpm", "duration": 30, "steps": 8},
    timeout=600,
)
r.raise_for_status()
print(r.json())

JavaScript:

const form = new FormData();
form.append("prompt", "warm analog synth arpeggio, 120 bpm");
form.append("duration", "30");
form.append("steps", "8");
const res = await fetch("http://localhost:8600/api/generate", { method: "POST", body: form });
console.log(await res.json());

Import audio into the library

curl:

curl -X POST http://localhost:8600/api/library/import \
  -F "file=@track.wav" \
  -F 'metadata={"title": "My Track", "source": "import"}'

Python:

import json
import requests

with open("track.wav", "rb") as f:
    r = requests.post(
        "http://localhost:8600/api/library/import",
        files={"file": f},
        data={"metadata": json.dumps({"title": "My Track", "source": "import"})},
    )
r.raise_for_status()
print(r.json()["id"])

JavaScript:

const form = new FormData();
form.append("file", fileBlob, "track.wav");
form.append("metadata", JSON.stringify({ title: "My Track", source: "import" }));
const res = await fetch("http://localhost:8600/api/library/import", { method: "POST", body: form });
console.log(await res.json());

The rest of the surface (stems, MIDI conversion, notation, DJ, VJ, project import, plugins) is browsable at /docs while the backend is running.