calcuis/hidream-gguf

gguf quantized version of HiDream-i1-Full (incl. full + dev + fast)

• full set gguf works right away (all gguf: model + encoder + vae)
• upgrade your node(pypi|repo|pack) for model support

setup

• drag hidream to > ./ComfyUI/models/diffusion_models
• drag encoder: g, l, t5xxl and llama to > ./ComfyUI/models/text_encoders
• drag pig to > ./ComfyUI/models/vae

workflow

• drag the json file or demo picture below for example workflow

Prompt

cute anime girl with massive fluffy fennec ears and a big fluffy tail blonde messy long hair blue eyes wearing a maid outfit with a long black gold leaf pattern dress and a white apron mouth open holding a fancy black forest cake with candles on top in the kitchen of an old dark Victorian mansion lit by candlelight with a bright window to the foggy forest and very expensive stuff everywhere

Negative Prompt

full (info only, don't copy this)

Prompt

fast (same prompt as full)

Prompt

dev (same prompt as full)

Prompt

1-clip (llama only)

Prompt

2-clip (llama + t5xxl)

reference

• base model from hidream-ai
• get safetensors from comfy-org
• get more t5xxl-encoder gguf here
• get more lama3.1-8b-encoder gguf here