This workflow is designed to be simple and easy to use. It perfectly fills the screen, so you don't have to move the UI around at all.
Requirements:
ComfyUI version: 0.3.29
GGUF Loader Node: https://github.com/city96/ComfyUI-GGUF
GGUF Models: https://huggingface.co/city96/HiDream-I1-Dev-gguf I could run the Q_8 model on my RTX 3080 10GB. For lower VRAM, please select Q_6, Q_4_K_M, ...
Text Encoders: https://huggingface.co/Comfy-Org/HiDream-I1_ComfyUI/tree/main/split_files/text_encoders
VAE: https://huggingface.co/HiDream-ai/HiDream-I1-Dev/blob/main/vae/diffusion_pytorch_model.safetensors or https://huggingface.co/black-forest-labs/FLUX.1-dev/blob/main/ae.safetensors
Recommended Settings:
Steps: 15-20
CFG: 1-2
Sampler: DDIM
Scheduler: ddim_uniform
Shift: 3-5
Ultimate Upscaler
Steps: 7-15
CFG: 1-2
Sampler: DDIM
Scheduler: ddim_uniform
Denoise: 0.15-0.4
Description
slight rearrangements
added upscaler
changed default values
FAQ
Comments (2)
I'm a bit confused over the upscaler. I have tried upscale by 2.00 and 4.00. Whatever I select, the upscaled image is the exact same dimensions as the original. Another issue related to the upscaler... the comparator node shows a difference between the original and upscale image but the actual saved imaged (HiDream_Upscaled_xxx and HiDream_xxx) are pixel exactly the same - I see zero difference.
THank you brother. Clean workflow working fine. I had to replace the upscaler for the Ultimate SD upscaler since the one on your W/F was giving me issues. Other than that, all good.
Looks like we don't have an active mirror for this file right now.
CivArchive is a community-maintained index — we catalog mirrors that volunteers upload to HuggingFace, torrents, and other public hosts. Looks like no one has uploaded a copy of this file yet.
Some files do get recovered over time through contributions. If you're looking for this one, feel free to ask in Discord, or help preserve it if you have a copy.