Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# Miscellaneous Features Models
|
| 2 |
+
|
| 3 |
+
Models for high-performance video/audio generation miscellaneous features.
|
| 4 |
+
|
| 5 |
+
## Features Supported
|
| 6 |
+
|
| 7 |
+
### Audio
|
| 8 |
+
- **ACE-Step**: AI music generation from prompts + lyrics
|
| 9 |
+
- **Podcast**: Multi-speaker podcast generation (SoulX-Podcast-1.7B)
|
| 10 |
+
|
| 11 |
+
### Video Processing
|
| 12 |
+
- **Face Enhancer**: Video face enhancement/swap (ReActor + CodeFormer/GPEN)
|
| 13 |
+
- **RIFE Upscaler**: Video frame interpolation + upscaling
|
| 14 |
+
- **Lucy Edit**: Text-guided video editing
|
| 15 |
+
- **Fusion I2V**: 4-image to video generation (Phantom)
|
| 16 |
+
- **Animate Posenet**: Character animation with pose control (WAN 2.2 Animate)
|
| 17 |
+
- **Camera Control**: Camera movement video generation
|
| 18 |
+
- **Animate Photo**: Motion transfer from video to image
|
| 19 |
+
|
| 20 |
+
## Model Categories
|
| 21 |
+
|
| 22 |
+
| Category | Description |
|
| 23 |
+
|----------|-------------|
|
| 24 |
+
| `diffusion_models/` | Main diffusion models (Lucy Edit, Phantom, Animate, Camera) |
|
| 25 |
+
| `text_encoders/` | UMT5-XXL text encoders |
|
| 26 |
+
| `vae/` | WAN VAE decoders |
|
| 27 |
+
| `clip_vision/` | CLIP Vision encoder |
|
| 28 |
+
| `loras/` | Lightning and LightX2V LoRAs |
|
| 29 |
+
| `facerestore_models/` | Face restoration models (CodeFormer, GPEN) |
|
| 30 |
+
| `insightface/` | Face detection and swap models |
|
| 31 |
+
| `upscale_models/` | RealESRGAN upscaler |
|
| 32 |
+
| `rife/` | RIFE frame interpolation |
|
| 33 |
+
| `sams/` | SAM2 segmentation |
|
| 34 |
+
| `onnx/` | ONNX detection models (VitPose, YOLOv10) |
|
| 35 |
+
| `ace_step/` | ACE-Step music generation models |
|
| 36 |
+
| `TTS/SoulX-Podcast-1.7B/` | Podcast generation models |
|
| 37 |
+
|
| 38 |
+
## Usage with RunPod
|
| 39 |
+
|
| 40 |
+
Set in RunPod endpoint settings:
|
| 41 |
+
- **Model**: `Aloukik21/unreal-misc`
|
| 42 |
+
- **HF Token**: Your HuggingFace token
|
| 43 |
+
|
| 44 |
+
The models will be automatically cached and used by the misc Docker container.
|