(Workflow)(Freebie)Hi-Dream ComfyUI Native Node Instruction And Workflow for txt2img With Upscaler (Patreon)

Tutorial Video : https://youtu.be/LyKhRvDWMRE

Files to Download

Download the text encoder files:

clip_l_hidream.safetensors

clip_g_hidream.safetensors

t5xxl_fp8_e4m3fn_scaled.safetensors

llama_3.1_8b_instruct_fp8_scaled.safetensors

Put these 4 files in your ComfyUI/models/text_encoders directory.

You can find them all: here. You might already have t5xxl downloaded.

The VAE can be found here and should go in your ComfyUI/models/vae/ folder. This is the Flux VAE so you might already have it.

The diffusion models can be found in this folder

For HiDream Dev:

Download hidream_i1_dev_bf16.safetensors and put it in your ComfyUI/models/diffusion_models/ directory.

For HiDream Full:

Download hidream_i1_full_fp16.safetensors and put it in your ComfyUI/models/diffusion_models/ directory.

Official sampling settings

Provided for reference

HiDream Full

* hidream_i1_full_fp16.safetensors

* shift: 3.0

* steps: 50

* sampler: uni_pc

* scheduler: simple

* cfg: 5.0

HiDream Dev

* hidream_i1_dev_bf16.safetensors

* shift: 6.0

* steps: 28

* sampler: lcm

* scheduler: normal

* cfg: 1.0 (no negative prompt)

HiDream Fast

* hidream_i1_fast_bf16.safetensors

* shift: 3.0

* steps: 16

* sampler: lcm

* scheduler: normal

* cfg: 1.0 (no negative prompt)

Via Comfy Examples

Attached the workflow that run in this tutorial video: