Sonora Studio: free local web UI for ACE-Step 1.5 music generation on ComfyUI
baxxox · reddit · 2026-10-11
Barbara released Sonora Studio, a free MIT-licensed local web UI that uses your own ComfyUI backend to run ACE-Step 1.5 music generation. It builds the ComfyUI graph via API from a style prompt and lyrics, and adds a song library, remix/repaint, audio editing, stem separation and mastering. Optional RunPod Serverless or Vast.ai backends for those without GPUs. Requires 12GB VRAM NVIDIA GPU, Node.js 22+, FFmpeg; model files (19.9GB) are SHA-256 verified. Windows quick start in the README.
More from Multimodal
- SeedVR2 Video Upscaler v1.6.6 Adds W4A8 Support: Better Quality Than NVFP4 at 4-bit Size — Zestyclose_Bake3680 · 2026-10-11
- Krea 2's weak prompts traced to a 256-token default and bare system prompt — Atmey · 2026-10-11
- AI-Generated Sci-Fi Series Cost $38,000 in Credits, 8-10 Generations per Shot — lulzxdxdxd · 2026-10-11
- One song, three AIs: lyrics by ChatGPT 5.2, music by Suno 4.5, video by Claude Opus 5.5 — technollama · 2026-10-11
- Quartermaster: open-source local server runs Qwen-Image 2.1 Turbo without ComfyUI — OneMoreName1 · 2026-10-11
- AI-generated track 'Grok, Is This True?' dubbed slopcore for the masses — technollama · 2026-10-11