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

Techno

Techno-style generation. Covers: driving techno, four on the floor kick, hypnotic synth pulse, deep bass; dark techno, pounding kick, distorted bass, atmospheric pads, hypnotic; and minimal techno, stripped percussion, subtle bass groove, glitchy textures.

Quick start

  • Base model: AceStep v1.5 Turbo (2B)
  • Trigger word: roti-t3kn0 (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

  • driving techno, four on the floor kick, hypnotic synth pulse, deep bass (128 BPM, G minor)
  • dark techno, pounding kick, distorted bass, atmospheric pads, hypnotic (130 BPM, C minor)
  • minimal techno, stripped percussion, subtle bass groove, glitchy textures (125 BPM, F 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, 4305s total, instrumental fraction 0.2

Files

  • techno-v1.safetensors โ€” PEFT LoRA weights (88,130,248 bytes, sha256 0658581b527c1e9cf031e1abbed5d159901d09a5b93e306574cb576dbf737a1d)
  • techno-v1.metadata.json โ€” machine-readable sidecar (schema v1)
  • adapter_config.json โ€” PEFT adapter config

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

Downloads last month
12
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support