Update README.md

README.md

@@ -1,3 +1,18 @@
 ---
+pipeline_tag: video-to-video
 license: cc-by-nc-4.0
 ---
+![model example](https://i.imgur.com/ze1DGOJ.png)
+[example outputs](https://www.youtube.com/watch?v=HO3APT_0UA4) (courtesy of [dotsimulate](https://www.instagram.com/dotsimulate/))
+
+# zeroscope_v2 1111 models
+A collection of watermark-free Modelscope-based video models capable of generating high quality video at [448x256](https://huggingface.co/cerspense/zeroscope_v2_dark_30x448x256), [576x320](https://huggingface.co/cerspense/zeroscope_v2_576w) and [1024 x 576](https://huggingface.co/cerspense/zeroscope_v2_XL). These models were trained from the [original weights](https://huggingface.co/damo-vilab/modelscope-damo-text-to-video-synthesis) with offset noise using 9,923 clips and 29,769 tagged frames.<br />
+This collection makes it easy to switch between models with the new dropdown menu in the 1111 extension.
+
+### Using it with the 1111 text2video extension
+Simply download the contents of this repo to 'stable-diffusion-webui\models\text2video'
+Or, manually download the model folders you want, along with VQGAN_autoencoder.pth.
+
+Thanks to [dotsimulate](https://www.instagram.com/dotsimulate/) for the config files.
+
+Thanks to [camenduru](https://github.com/camenduru), [kabachuha](https://github.com/kabachuha), [ExponentialML](https://github.com/ExponentialML), [VANYA](https://twitter.com/veryVANYA), [polyware](https://twitter.com/polyware_ai), [tin2tin](https://github.com/tin2tin)<br />