You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

YAML Metadata Warning:empty or missing yaml metadata in repo card

Check out the documentation for more information.

primitive-vace-49f checkpoints (private)

Wan2.1-VACE-1.3B finetunes for primitive-conditioned novel-view video synthesis, trained on the 49-frame dataset (wliu283/datasets_49).

Run: pvace-49f-full-blob2 (full finetune, freeze_mode=none), lr 2e-5 cosine, eff-batch 32, 49 frames, 832x480.
Step 6000 (run targets 17000).
Held-out quality (N=50): LPIPS 0.062, SSIM 0.898, PSNR 28.0 dB.
Base model: Wan-AI/Wan2.1-VACE-1.3B-Diffusers.
Format: full training_state.pt (transformer weights + optimizer + global_step). Load via the repo's primitive_vace/inference.py / resume path (is_lora=False -> full transformer state_dict).

Downloads last month: -; Downloads are not tracked for this model. How to track

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support