Commit 414c3cd9 authored by gushiqiao's avatar gushiqiao
Browse files

FIX

parent 2931a72e
......@@ -25,8 +25,8 @@ Model Root Directory/
│ ├── fp8/ # FP8 quantized version (DIT/T5/CLIP)
│ ├── int8/ # INT8 quantized version (DIT/T5/CLIP)
│ ├── original/ # Original precision version (DIT)
│ ├── xlm-roberta-large/ # Multilingual encoder
│ └── google/ # Other shared resources
│ ├── xlm-roberta-large/
│ └── google/
```
### 💾 Storage Recommendations
......@@ -40,20 +40,6 @@ Model Root Directory/
/opt/models/ # System optimization directory
```
## 🔧 Model File Description
### Core Model Files
Each model directory contains the following core files:
| Filename | Size | Purpose | Required |
|----------|------|---------|----------|
| `config.json` | ~250B | Model configuration file | ✅ Required |
| `Wan2.1_VAE.pth` | ~508MB | VAE variational autoencoder | ✅ Required |
| `models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth` | ~4.77GB | CLIP visual encoder (FP16) | ✅ Required |
| `models_t5_umt5-xxl-enc-bf16.pth` | ~11.4GB | T5 text encoder (BF16) | ✅ Required |
| `taew2_1.pth` | ~22.7MB | Lightweight VAE (optional) | ❌ Optional |
### Quantized Version Directories
Each model contains multiple quantized versions for different hardware configurations:
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment