State current generation speed in the model introduction
Browse files
README.md
CHANGED
|
@@ -9,7 +9,7 @@ tags:
|
|
| 9 |
|
| 10 |
# Agnes-Video-2.0-MoE
|
| 11 |
|
| 12 |
-
Agnes-Video-2.0-MoE is an audio-video generation model based on DaVinci-MagiHuman. It accepts text, video, and audio inputs and generates video and audio.
|
| 13 |
|
| 14 |
## Model Description
|
| 15 |
|
|
|
|
| 9 |
|
| 10 |
# Agnes-Video-2.0-MoE
|
| 11 |
|
| 12 |
+
Agnes-Video-2.0-MoE is an audio-video generation model based on DaVinci-MagiHuman. It accepts text, video, and audio inputs and generates video and audio. With TIRx kernels, a 5-second 480×272 clip at 32 denoising steps now takes **13.641 s** on an NVIDIA B200, **4.150 s faster** than 17.791 s, a **23.33%** reduction.
|
| 13 |
|
| 14 |
## Model Description
|
| 15 |
|