 |
| Hidream o1 showcase |
Installation
1. Make sure you do the ComfyUI installation. Older user need to update it from the Manager itself.
2. Download HiDream o1 models from Officially repacked by ComfyUI:
There are multiple variants to choose from. Use that's suitable for you system resources:
(a) hidream_o1_image_bf16.safetensors
(b) hidream_o1_image_dev_bf16.safetensors
(c) hidream_o1_image_dev_fp8_scaled.safetensors
(d) hidream_o1_image_dev_mxfp8.safetensors
(e) hidream_o1_image_fp8_scaled.safetensors
(f) hidream_o1_image_mxfp8.safetensors
mxfp8 is the hardware level support (on RTX 4090 and 5090) latest release by NVIDIA (blackwell) for better quality and faster generation.
Here, the hidream_o1_image is the base variant and hidream_o1_image dev is the distilled variant supports text to image and image to image. Save this into ComfyUI/models/checkpoints folder.
3. This model is the unified transfer model. There are no VAE and text encoders. All these are packed into single diffusion model.
4. Restart and refresh ComfyUI.
Workflow
1. Download the workflows from our Hugging face repository.
(a)HiDream O1 Image to Image (HiDream_O1_Img2Img.json)
(b)HiDream o1 Text to Image(HiDream_O1_Txt2Img.json)
2. Drag and drop into ComfyUI
3. Load the HiDream O1 model into load checkpoint node.
4. Add text prompts into prompt box.
5. Set KSampler Settings-
HiDream o1 Image
Steps- 50
HiDream o1 Image Dev
Steps-28
6. Hit run to start generation.
- Better Text generation){kind=link}