Skip to content

Style-skin v5 stack (teacher-regen LoRA + re-paired corrector, final serving config)

Choose a tag to compare

View all tags
@wenqingw-nv wenqingw-nv released this 19 Aug 02:13
style-skin-v5-stack
1705825
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

Final Issue-1 serving configuration for the omnidreams game-skin edit:

  • lora_style_v5_step1600.pt — r64 multi-style LoRA trained on teacher-regenerated (SDEdit sigma 0.65 through the 35-step bidirectional teacher) styled targets. Kills the v3 +10..+14 style melt; ~2x usable style-hold depth with a flat post-swap divergence profile.
  • lora_style_corrector_v5_valpeak.pt — rank-16 clean-forcing drift corrector re-paired against v5 (val dag-R^2 +0.368).
  • gate_style_v5.json — measured style-specific alpha gate profile (t803 = 0.995).

Serve with: EDIT_LORA= STYLE_CORRECTOR= CORRECTOR_GAIN=0.15 GATE_ALPHA_JSON=, plus unsharp post (1.8/-0.8, sigma 2). Known residual: faint sky banding from the corrector (RMS 0.26 vs 0.12 bare). Alt gain 0.20 trades sky cleanliness for deep-window sharpness. Evidence: PR NVIDIA#458.

Assets 5
Loading