ZhengmingYu commited on
Commit
2d5597b
·
verified ·
1 Parent(s): b2812d5

Model card: title, badges and teaser

Browse files
Files changed (3) hide show
  1. .gitattributes +1 -0
  2. README.md +17 -8
  3. assets/teaser.gif +3 -0
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ assets/teaser.gif filter=lfs diff=lfs merge=lfs -text
README.md CHANGED
@@ -13,14 +13,23 @@ tags:
13
  - minimax-h3
14
  ---
15
 
16
- # DMAD students of MiniMax-H3 (4 steps, text-to-audio-video)
17
-
18
- <!-- TODO: paper / project page / code links -->
19
- Two 4-step students of [MiniMax-H3](https://huggingface.co/MiniMaxAI/MiniMax-H3) (33B) distilled with **DMAD,
20
- Distribution Matching as Adversarial Distillation** ([paper](https://arxiv.org/abs/TODO),
21
- [project page](https://yzmblog.github.io/DMAD/), [code](https://github.com/TODO)). Each is a rank-128 LoRA on the
22
- H3 text-to-audio-video transformer that generates 1344x768 video with native stereo audio in 4 model evaluations
23
- instead of the teacher's 50.
 
 
 
 
 
 
 
 
 
24
 
25
  | File | Checkpoint | Size |
26
  |------|------------|------|
 
13
  - minimax-h3
14
  ---
15
 
16
+ # DMAD: Distribution Matching as Adversarial Distillation for Fast Visual Generation
17
+
18
+ ### 4-step MiniMax-H3 students for joint audio-video generation
19
+
20
+ <!-- TODO: paper and code links -->
21
+ [![Project Page](https://img.shields.io/badge/Project-Page-yellow?logo=googlechrome&logoColor=yellow)](https://yzmblog.github.io/DMAD/)
22
+ [![Paper](https://img.shields.io/badge/Paper-arXiv-b31b1b?logo=arxiv&logoColor=red)](https://arxiv.org/abs/TODO)
23
+ [![Code](https://img.shields.io/badge/Code-GitHub-black?logo=github)](https://github.com/TODO)
24
+
25
+ <p align="center">
26
+ <img src="https://huggingface.co/ZhengmingYu/DMAD/resolve/main/assets/teaser.gif" alt="Videos generated by the 4-step DMAD student of MiniMax-H3 (video only; every clip also has generated audio)">
27
+ </p>
28
+
29
+ This repository holds the **4-step DMAD students of [MiniMax-H3](https://huggingface.co/MiniMaxAI/MiniMax-H3)
30
+ (33B, text-to-audio-video)**: rank-128 LoRAs on the H3 transformer that turn the 50-step teacher into a 4-step
31
+ generator of 1344x768 video with native stereo audio. The inference code is in the
32
+ [code repository](https://github.com/TODO).
33
 
34
  | File | Checkpoint | Size |
35
  |------|------------|------|
assets/teaser.gif ADDED

Git LFS Details

  • SHA256: ff8c375e432e677369a78539c4a85846aff9afcce6529e669b9b5cc78a077822
  • Pointer size: 133 Bytes
  • Size of remote file: 15.3 MB