Download
1 variant available
bf16 Other
pid_flux1_512_to_2048_4step_INT8.safetensors
BF16, good balance • 1.29 GB
Verified: a month ago
820 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 9
(6)
May 30, 2026
Other
Works with Zimage/Turbo and Flux1 (or any flux1 vae model)

3770 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 9
License:
Color fix for the flux2 PiD models (flux2_1024_4096_2606)
INT8ConvRot quantizations for PiD decode upscale models
PiD reformulates the latent-to-pixel decoder as a conditional pixel-space diffusion model, unifying decoding and upsampling into a single generative module. It denoises directly in high-resolution pixel space and produces a super-resolved image in one pass. This repository hosts the released decoder checkpoints, plus the encoder/decoder ("VAE") weights they depend on.
BF16 models here https://huggingface.co/Comfy-Org/PixelDiT/tree/main/diffusion_models
Load your INT8 models with this custom node for speedup on RTX 30XX & 20XX series GPU https://github.com/BobJohnson24/ComfyUI-INT8-Fast

