Hugging Face's logo

madtune
/
pixeldit-diffusers

Text-to-Image
Diffusers
Safetensors
PixelDiTPipeline
diffusion
pixeldit
nvidia
pixel-space
lora
Model card Files
xet
 Community
madtune commited on
Commit
71d5c90
·
verified ·
1 Parent(s): 1b1e962

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +11 -0
README.md CHANGED
@@ -24,6 +24,17 @@ All credit for the model architecture and weights goes to NVIDIA Research. This
24
 
25
  ---
26
 
 
 
 
 
 
 
 
 
 
 
 
27
  ## What is PixelDiT?
28
 
29
  PixelDiT is a 1.3B parameter **pixel-space** diffusion transformer — no VAE, generates images directly in pixel space. Runs on **4GB VRAM**.
 
24
 
25
  ---
26
 
27
+ ## Gallery — IP-Adapter style transfer (SigLIP only, no text prompt)
28
+
29
+ | | | | | |
30
+ |---|---|---|---|---|
31
+ | ![](assets/gallery/01_cyber_soldier.jpg) | ![](assets/gallery/02_neon_goth.jpg) | ![](assets/gallery/03_demon_queen.jpg) | ![](assets/gallery/04_punk_girl.jpg) | ![](assets/gallery/05_void_entity.jpg) |
32
+ | ![](assets/gallery/06_rain_alley.jpg) | ![](assets/gallery/07_ink_mask.jpg) | ![](assets/gallery/08_retro_city.jpg) | ![](assets/gallery/09_dream_goddess.jpg) | ![](assets/gallery/10_cosmic_entity.jpg) |
33
+
34
+ > All generated with [`madtune/pixeldit-controlnet`](https://huggingface.co/madtune/pixeldit-controlnet) — IP-Adapter only, zero text conditioning.
35
+
36
+ ---
37
+
38
  ## What is PixelDiT?
39
 
40
  PixelDiT is a 1.3B parameter **pixel-space** diffusion transformer — no VAE, generates images directly in pixel space. Runs on **4GB VRAM**.