theluyuan/DiffSynth-Studio
1
0
Fork 0
mirror of https://github.com/modelscope/DiffSynth-Studio.git synced 2026-03-19 14:58:12 +00:00
Code Issues Packages Projects Releases Wiki Activity

update docs

Browse Source
This commit is contained in:
Artiprocher
2024-11-04 14:09:36 +08:00
parent deff4512f7
commit fe68a3d1bb
5 changed files with 15 additions and 70 deletions
Download Patch File Download Diff File Expand all files Collapse all files

43
docs/source/tutorial/ASimpleExample.md
View File

@@ -17,25 +17,19 @@ pip install -e .
通过运行以下代码，我们将会下载模型、加载模型、生成图像。
```python
import torch
from diffsynth import ModelManager, FluxImagePipeline
from diffsynth import ModelManager, SDXLImagePipeline
model_manager = ModelManager(
torch_dtype=torch.bfloat16,
device="cuda",
model_id_list=["FLUX.1-dev"]
)
pipe = FluxImagePipeline.from_model_manager(model_manager)
model_manager = ModelManager(device="cuda", model_id_list=["StableDiffusionXL_v1"])
pipe = SDXLImagePipeline.from_model_manager(model_manager)
torch.manual_seed(0)
image = pipe(
prompt="In a forest, a wooden plank sign reading DiffSynth",
height=576, width=1024,
prompt="Diffuse light particles in the universe",
height=576, width=1024, seed=0
)
image.save("image.jpg")
```
![image](https://github.com/user-attachments/assets/15a52a2b-2f18-46fe-810c-cb3ad2853919)
![image](https://github.com/user-attachments/assets/2e60d18e-534c-43d6-b875-26db5b05442e)
从这个例子中，我们可以看到，DiffSynth 中有两个关键模块：`ModelManager` 和 `Pipeline`，接下来我们详细介绍。
@@ -44,30 +38,19 @@ image.save("image.jpg")
`ModelManager` 负责下载和加载模型，通过以下代码可以直接一步完成。
```python
import torch
from diffsynth import ModelManager
from diffsynth import ModelManager, SDXLImagePipeline
model_manager = ModelManager(
torch_dtype=torch.bfloat16,
device="cuda",
model_id_list=["FLUX.1-dev"]
)
model_manager = ModelManager(device="cuda", model_id_list=["StableDiffusionXL_v1"])
```
当然，我们也支持分步完成，以下代码和上述代码的行为是等价的。
```python
import torch
from diffsynth import download_models, ModelManager
download_models(["FLUX.1-dev"])
model_manager = ModelManager(torch_dtype=torch.bfloat16, device="cuda")
model_manager.load_models([
"models/FLUX/FLUX.1-dev/text_encoder/model.safetensors",
"models/FLUX/FLUX.1-dev/text_encoder_2",
"models/FLUX/FLUX.1-dev/ae.safetensors",
"models/FLUX/FLUX.1-dev/flux1-dev.safetensors"
])
download_models(["StableDiffusionXL_v1"])
model_manager = ModelManager(device="cuda")
model_manager.load_models(["models/stable_diffusion_xl/sd_xl_base_1.0.safetensors"])
```
下载模型时，我们支持从 [ModelScope](https://www.modelscope.cn/) 和 [HuggingFace](https://huggingface.co/) 下载模型，也支持下载非预置的模型，关于模型下载的更多信息请参考[模型下载](./DownloadModels.md)。
@@ -76,10 +59,10 @@ model_manager.load_models([
## 构建 Pipeline
DiffSynth-Studio 提供了多个推理 `Pipeline`，这些 `Pipeline` 可以直接通过 `ModelManager` 获取所需的模型并初始化。例如，FLUX.1-dev 模型的文生图 `Pipeline` 可以这样构建：
DiffSynth-Studio 提供了多个推理 `Pipeline`，这些 `Pipeline` 可以直接通过 `ModelManager` 获取所需的模型并初始化。例如，Kolors（可图）模型的文生图 `Pipeline` 可以这样构建：
```python
pipe = FluxImagePipeline.from_model_manager(model_manager)
pipe = SDXLImagePipeline.from_model_manager(model_manager)
```
更多用于图像生成和视频生成的 `Pipeline` 详见[推理流水线](./Pipelines.md)。