A 1-click, 4-step upscaler built around NVIDIA's PiD (Pixel Diffusion) model on the Flux1 backbone. Feed it any image and it intelligently regenerates real, fine detail at 4× resolution — then supersamples it back down to a razor-sharp final image. An automatic captioner keeps the upscale faithful to your image's content, so you get crisp detail without hallucinated junk.
✨ What makes this workflow different
Works with ANY aspect ratio — 16:9, 9:16, square, anything. The input is auto-normalized to the model's native 1024px long-edge (e.g. a 1280×720 image becomes 1024×576), which is the #1 cause of broken/green-tinted PiD results. This is handled for you.
Auto-prompting via Florence-2 — a vision model reads your image and writes a detailed caption automatically, guiding the upscaler to preserve the actual content. No manual prompting needed.
Locked to the model's true 4× regime — PiD
1024→4096is a fixed 4× model; this workflow targets exactly that for maximum sharpness (no blur from under/over-scaling).Supersampled final output — the 4096px result is Lanczos-downscaled back to 1024px, baking all 4× of generated detail into a clean, antialiased image with zero quality loss. You also get the full 4096px version saved.
Side-by-side comparison — a built-in slider compares your original against the result.
Fast — only 4 sampling steps (LCM).
🔧 How to use
Install the 3 model files (links + folder layout below).
Load the workflow, drop your image into the Load Image node.
Hit Queue. That's it.
Outputs:
Full 4× image (e.g. 4096×2304) — saved via the first Save node.
Supersampled 1024 image — crisp, detail-packed, saved via the second Save node.
Tip: The PiD model is brightest/cleanest on near-square and landscape images. Extreme portrait crops can show mild color shifts — that's a known characteristic of the current distilled model, not the workflow.
📥 Downloads Links Below:
Gemma 2b: https://huggingface.co/Comfy-Org/PixelDiT/tree/main/text_encoders
PiD models: https://huggingface.co/Comfy-Org/PixelDiT/tree/main/diffusion_models
VAE: https://huggingface.co/Comfy-Org/z_image_turbo/tree/main/split_files/vae
📂 ComfyUI/
├── 📂 models/
│ ├── 📂 text_encoders/
│ │ └── gemma_2_2b_it_elm_bf16.safetensors
│ ├── 📂 vae/
│ │ └── ae.safetensors
│ └── 📂 diffusion_models/
│ └── pid_flux1_1024_to_4096_4step_bf16.safetensors
📋 Required custom nodes
ComfyUI-Florence2 (auto-captioning)
ComfyUI-Custom-Scripts (pythongosssss — ShowText / MathExpression)
ComfyUI-easy-use
rgthree-comfy (Image Comparer)
ComfyUI_Swwan (GetImageSizeAndCount)
Requires a recent ComfyUI build with PixelDiT / PiD support. ~16GB VRAM recommended for the full 4096px pass.