Ethanfel/ComfyUI-SelVA
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
76 Commits 6 Branches 0 Tags
614a2e02aa676e81ea29d61ca5e43a07d6a7be93
Commit Graph

4 Commits

Author SHA1 Message Date
Ethanfel 614a2e02aa fix: weights_only=False for SelVA checkpoints (PyTorch 2.6 compat) 
PyTorch 2.6 changed the default to weights_only=True. SelVA checkpoints
contain non-tensor types (numpy scalars etc.) that fail strict unpickling.
All weights come from trusted sources (jnwnlee/selva HF repo).

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
 2026-04-04 16:38:31 +02:00
Ethanfel 40388ba6de fix: negative_prompt inline (multiline:false) + VAE filename v1-44.pth not v1-44k.pth 
- SelvaSampler: multiline:false puts negative_prompt inline above sliders
- SelvaModelLoader: VAE filenames in download_utils are v1-16.pth/v1-44.pth,
  not v1-{mode}.pth (mode includes the 'k' suffix)

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
 2026-04-04 16:35:17 +02:00
Ethanfel 9a985499e7 feat: auto-download SelVA weights on first use 
Uses selva_core/utils/download_utils.py (already has URLs + MD5s for all
weights). Models download to models/selva/ on first load. Synchformer reuses
models/prismaudio/synchformer_state_dict.pth if already present (no duplicate
download for PrismAudio users), otherwise downloads to models/selva/.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
 2026-04-04 16:25:36 +02:00
Ethanfel fe94438356 feat: SelvaModelLoader node — loads TextSynch + MMAudio + FeaturesUtils 
Resolves weights from models/selva/. Reuses synchformer_state_dict.pth from
models/prismaudio/ (no duplicate download). Supports four variants:
small_16k / small_44k / medium_44k / large_44k.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
 2026-04-04 15:21:03 +02:00