theluyuan/DiffSynth-Studio
1
0
Fork 0
mirror of https://github.com/modelscope/DiffSynth-Studio.git synced 2026-03-23 17:38:10 +00:00
Code Issues Packages Projects Releases Wiki Activity

update examples

Browse Source
This commit is contained in:
Artiprocher
2024-09-10 17:36:35 +08:00
parent 76bd931d79
commit d242af8e22
6 changed files with 61 additions and 41 deletions
Download Patch File Download Diff File Expand all files Collapse all files

16
examples/image_synthesis/README.md
View File

@@ -85,19 +85,3 @@ We highly recommend you to use this model in the WebUI.
|"black car"|"red car"|
|-|-|
|![black_car](https://github.com/Artiprocher/DiffSynth-Studio/assets/35051019/7fbfd803-68d4-44f3-8713-8c925fec47d0)|![black_car_to_red_car](https://github.com/Artiprocher/DiffSynth-Studio/assets/35051019/aaf886e4-c33c-4fd8-98e2-29eef117ba00)|
### Example: Prompt Processing
If you are not native English user, we provide translation service for you. Our prompter can translate other language to English and refine it using "BeautifulPrompt" models. Please see [`sd_prompt_refining.py`](./sd_prompt_refining.py) for more details.
Prompt: "一个漂亮的女孩". The [translation model](https://huggingface.co/Helsinki-NLP/opus-mt-en-zh) will translate it to English.
|seed=0|seed=1|seed=2|seed=3|
|-|-|-|-|
|![0_](https://github.com/Artiprocher/DiffSynth-Studio/assets/35051019/ebb25ca8-7ce1-4d9e-8081-59a867c70c4d)|![1_](https://github.com/Artiprocher/DiffSynth-Studio/assets/35051019/a7e79853-3c1a-471a-9c58-c209ec4b76dd)|![2_](https://github.com/Artiprocher/DiffSynth-Studio/assets/35051019/a292b959-a121-481f-b79c-61cc3346f810)|![3_](https://github.com/Artiprocher/DiffSynth-Studio/assets/35051019/1c19b54e-5a6f-4d48-960b-a7b2b149bb4c)|
Prompt: "一个漂亮的女孩". The [translation model](https://huggingface.co/Helsinki-NLP/opus-mt-en-zh) will translate it to English. Then the [refining model](https://huggingface.co/alibaba-pai/pai-bloom-1b1-text2prompt-sd) will refine the translated prompt for better visual quality.
|seed=0|seed=1|seed=2|seed=3|
|-|-|-|-|
|![0](https://github.com/Artiprocher/DiffSynth-Studio/assets/35051019/778b1bd9-44e0-46ac-a99c-712b3fc9aaa4)|![1](https://github.com/Artiprocher/DiffSynth-Studio/assets/35051019/c03479b8-2082-4c6e-8e1c-3582b98686f6)|![2](https://github.com/Artiprocher/DiffSynth-Studio/assets/35051019/edb33d21-3288-4a55-96ca-a4bfe1b50b00)|![3](https://github.com/Artiprocher/DiffSynth-Studio/assets/35051019/7848cfc1-cad5-4848-8373-41d24e98e584)|

35
examples/image_synthesis/omost_flux_text_to_image.py
View File

@@ -1,35 +0,0 @@
import torch
from diffsynth import download_models, ModelManager, OmostPromter, FluxImagePipeline
download_models(["OmostPrompt"])
download_models(["FLUX.1-dev"])
model_manager = ModelManager(torch_dtype=torch.bfloat16)
model_manager.load_models([
"models/OmostPrompt/omost-llama-3-8b-4bits",
"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"
])
pipe_omost = FluxImagePipeline.from_model_manager(model_manager, prompt_extender_classes=[OmostPromter])
pipe = FluxImagePipeline.from_model_manager(model_manager)
prompt = "A witch uses ice magic to fight against wild beasts"
seed = 7
torch.manual_seed(seed)
image = pipe_omost(
prompt=prompt,
num_inference_steps=30, embedded_guidance=3.5
)
image.save(f"image_omost.jpg")
torch.manual_seed(seed)
image2= pipe(
prompt=prompt,
num_inference_steps=30, embedded_guidance=3.5
)
image2.save(f"image.jpg")

25
examples/image_synthesis/qwen_prompt_refining.py
View File

@@ -1,25 +0,0 @@
from diffsynth import ModelManager, SDXLImagePipeline, download_models, QwenPrompt
import torch
download_models(["StableDiffusionXL_v1", "QwenPrompt"])
# Load models
model_manager = ModelManager(torch_dtype=torch.float16, device="cuda")
model_manager.load_models([
"models/stable_diffusion_xl/sd_xl_base_1.0.safetensors",
"models/QwenPrompt/qwen2-1.5b-instruct",
])
pipe = SDXLImagePipeline.from_model_manager(model_manager, prompt_refiner_classes=[QwenPrompt])
prompt = "一个漂亮的女孩"
negative_prompt = ""
for seed in range(4):
torch.manual_seed(seed)
image = pipe(
prompt=prompt, negative_prompt=negative_prompt,
height=1024, width=1024,
num_inference_steps=30
)
image.save(f"{seed}.jpg")

30
examples/image_synthesis/sd_prompt_refining.py
View File

@@ -1,30 +0,0 @@
from diffsynth import ModelManager, SDXLImagePipeline, download_models, Translator, BeautifulPrompt
import torch
# Download models (automatically)
# `models/stable_diffusion_xl/sd_xl_base_1.0.safetensors`: [link](https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0/resolve/main/sd_xl_base_1.0.safetensors)
# `models/BeautifulPrompt/pai-bloom-1b1-text2prompt-sd/`: [link](https://huggingface.co/alibaba-pai/pai-bloom-1b1-text2prompt-sd)
# `models/translator/opus-mt-zh-en/`: [link](https://huggingface.co/Helsinki-NLP/opus-mt-en-zh)
download_models(["StableDiffusionXL_v1", "BeautifulPrompt", "opus-mt-zh-en"])
# Load models
model_manager = ModelManager(torch_dtype=torch.float16, device="cuda")
model_manager.load_models([
"models/stable_diffusion_xl/sd_xl_base_1.0.safetensors",
"models/BeautifulPrompt/pai-bloom-1b1-text2prompt-sd",
"models/translator/opus-mt-zh-en"
])
pipe = SDXLImagePipeline.from_model_manager(model_manager, prompt_refiner_classes=[Translator, BeautifulPrompt])
prompt = "一个漂亮的女孩"
negative_prompt = ""
for seed in range(4):
torch.manual_seed(seed)
image = pipe(
prompt=prompt, negative_prompt=negative_prompt,
height=1024, width=1024,
num_inference_steps=30
)
image.save(f"{seed}.jpg")