|
Download README.md from Avdpro/LivePortrait-CUDA: direct link, hf CLI and curl.
- Browser
- Download file 1.22 kB
-
https://huggingface.co/Avdpro/LivePortrait-CUDA/resolve/main/README.md
- Command line
-
hf download hf://Avdpro/LivePortrait-CUDA/README.md
-
curl -L -o README.md https://huggingface.co/Avdpro/LivePortrait-CUDA/resolve/main/README.md
1.22 kB
| license: mit | |
| library_name: pytorch | |
| pipeline_tag: image-to-video | |
| # AI2Apps LivePortrait CUDA checkpoint candidate | |
| Human portrait animation weights for the AI2Apps CUDA Worker. Seven lossless | |
| safetensors conversions, unchanged LivePortrait203 landmarks and unchanged | |
| YuNet detection. No InsightFace checkpoint is included. Preserve LICENSE, | |
| YUNET_LICENSE and NOTICE.md. | |
| Original weights: KlingTeam/LivePortrait revision | |
| 82a4fa6735ca58432b6ce39301b4b9ee066dea47. Architecture and configuration: | |
| KlingAIResearch/LivePortrait commit 9b294b3d0536135442ea73cb01e6cb3ca7029dd3. | |
| conversion-receipt.json records strict loading and lossless tensor roundtrips; | |
| original spectral-normalization tensors are retained. | |
| YuNet: opencv/opencv_zoo revision 47534e27c9851bb1128ccc0102f1145e27f23f98, | |
| models/face_detection_yunet/face_detection_yunet_2023mar.onnx. | |
| The checkpoint manifest binds all selected file sizes and SHA-256 digests. | |
| This is a distribution candidate. A published CUDA model Package, immutable | |
| dual-source verification and signed installation acceptance remain required. | |
| This card does not claim production release or broad quality certification. | |
| FP32 is the reference mode; BF16 is an explicit fast mode. | |