Upload README.md
Browse files- models/README.md +33 -0
models/README.md
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# Download Pretrained Models
|
| 2 |
+
|
| 3 |
+
All models are stored in `HunyuanCustom/models` by default, and the file structure is as follows
|
| 4 |
+
```shell
|
| 5 |
+
HunyuanCustom
|
| 6 |
+
├──models
|
| 7 |
+
│ ├──README.md
|
| 8 |
+
│ ├──hunyuancustom_720P
|
| 9 |
+
│ │ ├──mp_rank_00_model_states.pt
|
| 10 |
+
│ │ │──mp_rank_00_model_states_fp8.pt
|
| 11 |
+
│ │ ├──mp_rank_00_model_states_fp8_map.pt
|
| 12 |
+
├ ├──vae_3d
|
| 13 |
+
│ ├──openai_clip-vit-large-patch14
|
| 14 |
+
│ ├──llava-llama-3-8b-v1_1
|
| 15 |
+
├──...
|
| 16 |
+
```
|
| 17 |
+
|
| 18 |
+
## Download HunyuanCustom model
|
| 19 |
+
To download the HunyuanCustom model, first install the huggingface-cli. (Detailed instructions are available [here](https://huggingface.co/docs/huggingface_hub/guides/cli).)
|
| 20 |
+
|
| 21 |
+
```shell
|
| 22 |
+
python -m pip install "huggingface_hub[cli]"
|
| 23 |
+
```
|
| 24 |
+
|
| 25 |
+
Then download the model using the following commands:
|
| 26 |
+
|
| 27 |
+
```shell
|
| 28 |
+
# Switch to the directory named 'HunyuanCustom'
|
| 29 |
+
cd HunyuanCustom
|
| 30 |
+
# Use the huggingface-cli tool to download HunyuanCustom model in HunyuanCustom/models dir.
|
| 31 |
+
# The download time may vary from 10 minutes to 1 hour depending on network conditions.
|
| 32 |
+
huggingface-cli download tencent/HunyuanCustom --local-dir ./
|
| 33 |
+
```
|