VCLab-PolyU/PVD

Source: 🤗text-to-imageapache-2.062 GBsafetensors✓ 17 checksumsupdated 40d ago

CommentsShare your thoughts and shape the future of sovereign AI.

@VCLab-PolyU is unclaimedSign in with the matching Hugging Face account or an organization member account to verify this handle.Claim @VCLab-PolyU

Files and sources

SourceAvailable on 🤗

Checksums17 awaiting witness

TorrentNot listed

SwarmNot checked

Source revision: c7a3a5b9d2b0…
Witnessed: Not yet
Weight files: 17
Model size: 62 GB
Downloads · last month: 0
Publisher: Unclaimed

Public checksum record coming on the next sync

Weights

Original

Weights62 GB
17 files
weights/pvd_qwenimage/backbone.safetensors 20 GBweights/pvd_flux/backbone.safetensors 12 GBweights/pvd_sd35m_unsplash/part1.safetensors 5.4 GBweights/pvd_sd35m_unsplash/part2.safetensors 5.4 GBweights/pvd_sd35m/part1.safetensors 5.4 GBweights/pvd_sd35m/part2.safetensors 5.4 GBweights/pvd_c2i/model.safetensors 2.7 GBweights/pvd_flux_unsplash/part1/adapter_model.safetensors 680 MBweights/pvd_flux_unsplash/part2/adapter_model.safetensors 680 MBweights/pvd_qwen_unsplash/part1/adapter_model.safetensors 613 MBweights/pvd_qwen_unsplash/part2/adapter_model.safetensors 613 MBweights/pvd_qwenimage/part1/adapter_model.safetensors 613 MBweights/pvd_qwenimage/part2/adapter_model.safetensors 613 MBweights/pvd_flux/part1/adapter_model.safetensors 595 MBweights/pvd_flux/part2/adapter_model.safetensors 595 MBcache/vavae-imagenet256-f16d32-dinov2.pt 279 MBcache/latents_stats.pt 1.8 kB

Two Halves are More than One: Phase-wise Velocity Distillation for Fast and High-Quality Image Generation

Model weights for PVD class-conditioned image generation (C2I) and text-to-image generation with SD3.5 Medium, FLUX.1-dev and Qwen-Image.

Model Details

Variant	Weight directory	Released components	Configuration
C2I	`weights/pvd_c2i/`	`model.safetensors`; VAE and latent statistics in `cache/`	ImageNet classes, 256×256
SD3.5 Medium	`weights/pvd_sd35m/`	`part1.safetensors`, `part2.safetensors`, `model_config.json`	12 layers per transformer
SD3.5 Medium (Unsplash)	`weights/pvd_sd35m_unsplash/`	`part1.safetensors`, `part2.safetensors`, `model_config.json`	Further trained on the Unsplash dataset
FLUX.1-dev	`weights/pvd_flux/`	`backbone.safetensors`, `part1/` and `part2/` LoRA adapters	9 dual-stream + 19 single-stream layers; rank 64
FLUX.1-dev (Unsplash)	`weights/pvd_flux_unsplash/`	`part1/` and `part2/` LoRA adapters; reuses `weights/pvd_flux/backbone.safetensors`	Further trained on the Unsplash dataset
Qwen-Image	`weights/pvd_qwenimage/`	`backbone.safetensors`, `part1/` and `part2/` LoRA adapters	30 layers; rank 32
Qwen-Image (Unsplash)	`weights/pvd_qwen_unsplash/`	`part1/` and `part2/` LoRA adapters; reuses `weights/pvd_qwenimage/backbone.safetensors`	Further trained on the Unsplash dataset

Each adapter folder contains adapter_model.safetensors and adapter_config.json. The C2I decoder and latent statistics are cache/vavae-imagenet256-f16d32-dinov2.pt and cache/latents_stats.pt.

Usage

Place this repository's weights/ and cache/ folders in the PVD code repository root. Keep the folder names as listed above. See the code repository's README for base model components and inference commands.