ai model report
Is Wan-AI/Wan2.1-VACE-14B safe to load?
CLEAN risk. 5 of 35 tracked dependencies carry advisories. Driven by No version-matched advisories and no malicious packages..
not scoredno advisory applies to a version this model pinsCLEAN
LOWShips safetensors plus 2 pickle checkpoint(s) for other tooling (Wan2.1_VAE.pth, models_t5_umt5-xxl-enc-bf16.pth); from_pretrained never loads these, even with use_safetensors=False, but code that torch.load()s them directly (a reference implementation, a training script) executes pickle.
2 more findings on this model.
- Author
- Wan-AI
- Task
- image-to-video
- Loader
- diffusers
- License
- apache-2.0
- Downloads
- 75,565
- Remote code
- not required
7 direct, 6 with findings, 29 reachable
| Package | Evidence | Advisories | Worst |
|---|---|---|---|
| transformers | direct, inferred | 24 | none pinned |
| torch | direct, inferred | 12 | none pinned |
Create a free accountto see all 35 dependencies, which versions are pinned, and what to change.
Risk computed 2026-10-14 at 02:43 UTC. Dependencies come from the model's requirements, its declared loader and the code it ships; advisories from NVD, GHSA and OSV.