Configuration Parsing Warning:In adapter_config.json: "peft.task_type" must be a string

Lo-Fi

Lo-Fi-style generation. Covers: lo-fi hip hop beat, vinyl crackle, jazzy chords, mellow saxophone, chill; lo-fi chillhop, soft drums, warm electric piano, vinyl noise, sleepy; and lo-fi study music, mellow chords, soft drums, ambient textures.

Quick start

  • Base model: AceStep v1.5 Turbo (2B)
  • Trigger word: roti-l0f1y (must appear in the prompt to activate)
  • Recommended strength: 1.0
  • Tested strength range: 0.3 โ€“ 1.0
  • Recommended steps: 8
  • Recommended guidance: 7.0
  • Recommended shift: 3.0
  • Sample rate: 48000 Hz

Example prompts

  • lo-fi hip hop beat, vinyl crackle, jazzy chords, mellow saxophone, chill (80 BPM, F minor)
  • lo-fi chillhop, soft drums, warm electric piano, vinyl noise, sleepy (75 BPM, C major)
  • lo-fi study music, mellow chords, soft drums, ambient textures (70 BPM, A minor)

Training

  • Rank 64, alpha 128, dropout 0.1
  • Target modules: q_proj, k_proj, v_proj, o_proj
  • 500 epochs / 7500 steps, batch 1, optimizer adamw_fused, lr 0.0003, warmup 100
  • Dataset: 15 synthetic tracks, 3122s total, instrumental fraction 0.0

Files

  • lo_fi-v1.safetensors โ€” PEFT LoRA weights (88,130,248 bytes, sha256 d3c66b189844dc078ea0d583e842edde12c607c15d126f359bb6f506b6ef9dd1)
  • lo_fi-v1.metadata.json โ€” machine-readable sidecar (schema v1)
  • adapter_config.json โ€” PEFT adapter config

Full metadata: see lo_fi-v1.metadata.json.

Downloads last month
8
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support