由于AuraFlow模型比较大，我就下在本地/hf_hub,结果运行Huggingface上README.md的代码
from diffusers import AuraFlowPipeline
import torch
pipeline = AuraFlowPipeline.from_pretrained(
"/hf_hub/fal/AuraFlow-v0.2",
torch_dtype=torch.float16,
variant="fp16",
).to("cuda")
image = pipeline(
prompt="close-up portrait of a majestic iguana with vibrant blue-green scales, piercing amber eyes, and orange spiky crest. Intricate textures and details visible on scaly skin. Wrapped in dark hood, giving regal appearance. Dramatic lighting against black background. Hyper-realistic, high-resolution image showcasing the reptile's expressive features and coloration.",
height=1024,
width=1024,
num_inference_steps=50,
generator=torch.Generator().manual_seed(666),
guidance_scale=3.5,
).images[0]
image
ValueError: /hf_hub/fal/AuraFlow-v0.2/transformer/ containing more than one
.index.jsonfile, delete the irrelevant ones.
网上并没有资料，我只好做了些尝试，发现是transformer文件夹下index.json重合了。我就用link做了测试
解决方法
删除其中一个index.json。可参考我的目录
fal/
├── AuraFlow-v0.2
│ ├── aura_flow_0.2.safetensors -> /hf_hub/fal/AuraFlow-v0.2/aura_flow_0.2.safetensors
│ ├── model_index.json -> /hf_hub/fal/AuraFlow-v0.2/model_index.json
│ ├── scheduler -> /hf_hub/fal/AuraFlow-v0.2/scheduler
│ ├── text_encoder -> /hf_hub/fal/AuraFlow-v0.2/text_encoder/
│ ├── tokenizer -> /hf_hub/fal/AuraFlow-v0.2/tokenizer/
│ ├── transformer
│ │ ├── config.json -> /hf_hub/fal/AuraFlow-v0.2/transformer/config.json
│ │ ├── diffusion_pytorch_model-00001-of-00003.safetensors -> /hf_hub/fal/AuraFlow-v0.2/transformer/diffusion_pytorch_model-00001-of-00003.safetensors
│ │ ├── diffusion_pytorch_model-00002-of-00003.safetensors -> /hf_hub/fal/AuraFlow-v0.2/transformer/diffusion_pytorch_model-00002-of-00003.safetensors
│ │ ├── diffusion_pytorch_model-00003-of-00003.safetensors -> /hf_hub/fal/AuraFlow-v0.2/transformer/diffusion_pytorch_model-00003-of-00003.safetensors
│ │ └── diffusion_pytorch_model.safetensors.index.json -> /hf_hub/fal/AuraFlow-v0.2/transformer/diffusion_pytorch_model.safetensors.index.json
│ └── vae -> /hf_hub/fal/AuraFlow-v0.2/vae/
└── AuraFlow-v0.2-fp16
├── aura_flow_0.2.safetensors -> /hf_hub/fal/AuraFlow-v0.2/aura_flow_0.2.safetensors
├── model_index.json -> /hf_hub/fal/AuraFlow-v0.2/model_index.json
├── scheduler -> /hf_hub/fal/AuraFlow-v0.2/scheduler
├── text_encoder -> /hf_hub/fal/AuraFlow-v0.2/text_encoder/
├── tokenizer -> /hf_hub/fal/AuraFlow-v0.2/tokenizer/
├── transformer
│ ├── config.json -> /hf_hub/fal/AuraFlow-v0.2/transformer/config.json
│ ├── diffusion_pytorch_model-00001-of-00002.fp16.safetensors -> /hf_hub/fal/AuraFlow-v0.2/transformer/diffusion_pytorch_model-00001-of-00002.fp16.safetensors
│ ├── diffusion_pytorch_model-00002-of-00002.fp16.safetensors -> /hf_hub/fal/AuraFlow-v0.2/transformer/diffusion_pytorch_model-00002-of-00002.fp16.safetensors
│ └── diffusion_pytorch_model.safetensors.fp16.index.json -> /hf_hub/fal/AuraFlow-v0.2/transformer/diffusion_pytorch_model.safetensors.fp16.index.json
└── vae -> /hf_hub/fal/AuraFlow-v0.2/vae/
fp16
Fri Aug 2 19:48:23 2024
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 555.58.02 Driver Version: 555.58.02 CUDA Version: 12.5 |
|-----------------------------------------+------------------------+----------------------+
| GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. |
| | | MIG M. |
|=========================================+========================+======================|
| 0 NVIDIA GeForce RTX 4090 Off | 00000000:01:00.0 Off | Off |
| 34% 58C P0 407W / 515W | 18542MiB / 24564MiB | 100% Default |
| | | N/A |
+-----------------------------------------+------------------------+----------------------+
+-----------------------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=========================================================================================|
| 0 N/A N/A 1124 G /usr/lib/Xorg 167MiB |
| 0 N/A N/A 1189 G /usr/bin/sddm-greeter-qt6 146MiB |
| 0 N/A N/A 3878 C ...conda3/envs/ai-train/bin/python3.10 18196MiB |
+-----------------------------------------------------------------------------------------+
无fp16
Fri Aug 2 19:50:18 2024
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 555.58.02 Driver Version: 555.58.02 CUDA Version: 12.5 |
|-----------------------------------------+------------------------+----------------------+
| GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. |
| | | MIG M. |
|=========================================+========================+======================|
| 0 NVIDIA GeForce RTX 4090 Off | 00000000:01:00.0 Off | Off |
| 36% 44C P0 71W / 515W | 20834MiB / 24564MiB | 0% Default |
| | | N/A |
+-----------------------------------------+------------------------+----------------------+
+-----------------------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=========================================================================================|
| 0 N/A N/A 1124 G /usr/lib/Xorg 167MiB |
| 0 N/A N/A 1189 G /usr/bin/sddm-greeter-qt6 146MiB |
| 0 N/A N/A 4031 C ...conda3/envs/ai-train/bin/python3.10 20488MiB |
+-----------------------------------------------------------------------------------------+