Instructions to use ZhengmingYu/DMAD with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use ZhengmingYu/DMAD with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline from diffusers.utils import export_to_video # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("MiniMaxAI/MiniMax-H3", dtype=torch.bfloat16, device_map="cuda") pipe.load_lora_weights("ZhengmingYu/DMAD") prompt = "A man with short gray hair plays a red electric guitar." output = pipe(prompt=prompt).frames[0] export_to_video(output, "output.mp4") - Inference
- Notebooks
- Google Colab
- Kaggle
Commit History
18 days ago
-
Update README.md
Verified signatureb08c2bf -
Model card: sync the 8-step checkpoint row with the GitHub README
Verified signature700b984 -
Model card: the 8-step student 8step_full_critic
Verified signatured3f472a -
Add the 8-step MiniMax-H3 student (8step_full_critic) and its ComfyUI-layout LoRA
Verified signature4d5696d
-
Link the Wan2.1 training data (DMAD-Wan-data)
Verified signature439136a
-
Model card: BibTeX as @article (arXiv preprint 2610.02188)
Verified signature4e61167 -
Add ModelScope model links to DMAD model card
Verified signature00cab6b
-
Model card: library_name minimax-h3 (download counting covers the subfolder safetensors) + diffusers tag
Verified signature2153aed -
Model card: Updates section
Verified signature9a2d83d -
ComfyUI workflow for 8 GB GPUs (5 s); pick-by-GPU table; --low-vram on 8 GB / GPU-size presets
Verified signatureAdds minimax_h3/workflows/dmad_h3_4step_5s_8gb_guitar.{json,mp4} and its screenshot; model card: 8 GB section, workflow-by-GPU table, inference.py --low-vram on 8 GB GPUs.ad7675c
-
Model card: ComfyUI workflows for 16 GB GPUs and up (14 GiB-cap rows in the table)
Verified signaturebcfb2f9
-
Model card: 16 GB / 22 GB GPUs (ComfyUI workflows at a 14 GiB cap, inference.py --low-vram)
Verified signaturee66d813
-
Add ready-to-run ComfyUI workflows (4- and 8-step, 15 s, 24 GB GPU)
Verified signature1c16b8c
-
Move the ComfyUI-layout LoRAs into minimax_h3/ next to the Diffusers-layout ones; update the card
Verified signature233f4e2 -
Model card: ComfyUI sampling with stock lcm + simple (equivalent to the DMAD nodes)
Verified signaturec7340fd -
Model card: ComfyUI section
Verified signatureb1fa1f8 -
Add the ComfyUI-layout MiniMax-H3 LoRAs (exact conversions, rank 128)
Verified signaturee9a0d1a -
Model card: all released DMAD weights (MiniMax-H3, Wan2.1, SDXL, ImageNet-64) and per-family licenses
Verified signature6faf8fd -
Add the sdxl DMAD weights
Verified signature7deffdd -
Add the wan2.1 DMAD weights
Verified signature8b7f9e4 -
Add the imagenet64 DMAD weights
Verified signaturebf9236a
-
Model card: authors, demo video badge, citation
Verified signature723954a -
Model card: say where the inference code is and add the clone step
Verified signature510ddb6 -
Model card: project page URL
Verified signaturead6eafc -
Model card: link the arXiv paper
Verified signature3189bef -
Model card: link the code repository
Verified signature060492f -
Model card: title, badges and teaser
Verified signature2d5597b -
Model card: shorten the license paragraph
Verified signatureb2812d5 -
Add the 4-step DMAD students of MiniMax-H3 (lora_critic, full_critic), model card, license and notice
Verified signature2cb62f3 -
initial commit
Verified signature7ac099e