Upload 23 files

Dockerfile

@@ -57,8 +57,6 @@ RUN pyenv install $PYTHON_VERSION && \
 COPY --chown=user:user . .
 # change dir since pip needs to seed whl folder
 RUN cd server && pip install --no-cache-dir --upgrade -r requirements.txt 
-RUN --mount=type=secret,id=GITHUB_TOKEN,mode=0444,required=true \
-    pip install git+https://$(cat /run/secrets/GITHUB_TOKEN)@github.com/cumulo-autumn/StreamDiffusion.git
 RUN python -m streamdiffusion.tools.install-tensorrt
 
 COPY --from=frontend /app/build ./view/build

README-ja.md

@@ -0,0 +1,31 @@
+# Txt2Img Example
+
+[English](./README.md) | [日本語](./README-ja.md)
+
+<p align="center">
+  <img src="../../assets/demo_01.gif" width=80%>
+</p>
+
+StreamDiffusion を用いた GUI を提供します。
+入力プロンプトを変更すると、テキストから 4x4 の画像をリアルタイムに生成することができます。
+
+## 使用方法
+
+以下のコマンドを順番に実行してください。
+
+```bash
+pip install -r requirements.txt
+cd view
+npm i
+npm start &
+cd ../server
+python main.py
+```
+
+# 謝辞
+
+この GitHubリポジトリ にある動画と画像のデモは、[kohakuV2](https://civitai.com/models/136268/kohaku-v2)と[SD-Turbo](https://arxiv.org/abs/2311.17042)を使用して生成されました。
+
+KohakuV2 モデルを提供していただいたKohaku BlueLeaf 様 ([@KBlueleaf](https://twitter.com/KBlueleaf))、[SD-Turbo](https://arxiv.org/abs/2311.17042)を提供していただいた[Stability AI](https://ja.stability.ai/)様に心より感謝いたします。
+
+KohakuV2 モデルは [Civitai](https://civitai.com/models/136268/kohaku-v2) と [Hugging Face](https://huggingface.co/stabilityai/sd-turbo) からダウンロードでき、[SD-Turbo](https://arxiv.org/abs/2311.17042) は Hugging Faceで使用可能です。

requirements.txt

@@ -0,0 +1,2 @@
+uvicorn>=0.24.0
+fastapi>=0.104.1

server/config.py

@@ -26,9 +26,9 @@ class Config:
     ####################################################################
     mode: Literal["txt2img", "img2img"] = "txt2img"
     # SD1.x variant model
-    model_id: str = "stabilityai/sd-turbo"
+    model_id_or_path: str = "KBlueLeaf/kohaku-v2.1"
     # LCM-LORA model
-    lcm_lora_id: str = None
+    lcm_lora_id: str = "latent-consistency/lcm-lora-sdv1-5"
     # TinyVAE model
     vae_id: str = "madebyollin/taesd"
     # Device to use
@@ -36,7 +36,7 @@ class Config:
     # Data type
     dtype: torch.dtype = torch.float16
     # acceleration
-    acceleration: Literal["none", "xformers", "sfast", "tensorrt"] = "sfast"
+    acceleration: Literal["none", "xformers", "tensorrt"] = "xformers"
 
     ####################################################################
     # Inference configuration

server/main.py

@@ -1,3 +1,5 @@
+import sys
+import os
 import asyncio
 import base64
 import logging
@@ -12,8 +14,10 @@ from fastapi.staticfiles import StaticFiles
 
 from PIL import Image
 from pydantic import BaseModel
-from wrapper import StreamDiffusionWrapper
 
+sys.path.append(os.path.join(os.path.dirname(__file__), "..", "..", ".."))
+
+from utils.wrapper import StreamDiffusionWrapper
 
 logger = logging.getLogger("uvicorn")
 PROJECT_DIR = Path(__file__).parent.parent
@@ -56,7 +60,7 @@ class Api:
         self.config = config
         self.stream_diffusion = StreamDiffusionWrapper(
             mode=config.mode,
-            model_id=config.model_id,
+            model_id_or_path=config.model_id_or_path,
             lcm_lora_id=config.lcm_lora_id,
             vae_id=config.vae_id,
             device=config.device,
@@ -65,6 +69,7 @@ class Api:
             t_index_list=config.t_index_list,
             warmup=config.warmup,
             use_safety_checker=config.use_safety_checker,
+            cfg_type="none",
         )
         self.app = FastAPI()
         self.app.add_api_route(
@@ -103,7 +108,9 @@ class Api:
         """
         async with self._predict_lock:
             return PredictResponseModel(
-                base64_image=self._pil_to_base64(self.stream_diffusion(prompt=inp.prompt))
+                base64_image=self._pil_to_base64(
+                    self.stream_diffusion(prompt=inp.prompt)
+                )
             )
 
     def _pil_to_base64(self, image: Image.Image, format: str = "JPEG") -> bytes:

server/requirements.txt

@@ -2,7 +2,7 @@ xformers
 uvicorn[standard]==0.24.0.post1
 fastapi==0.104
 accelerate
-# git+https://github.com/cumulo-autumn/StreamDiffusion.git@main#egg=stream-diffusion
+git+https://github.com/cumulo-autumn/StreamDiffusion.git@main#egg="streamdiffusion[tensorrt]"
 --extra-index-url https://download.pytorch.org/whl/cu121
 torch 
 torchvision

start.sh

@@ -1,4 +1,4 @@
 #!/bin/bash
 pip install -r requirements.txt 
-cd view && npm run build && cd ..
+cd view && npm install && npm run build && cd ..
 cd server && python3 main.py

view/build/asset-manifest.json

@@ -0,0 +1,15 @@
+{
+  "files": {
+    "main.css": "/static/css/main.ee59dfc2.css",
+    "main.js": "/static/js/main.d05767f4.js",
+    "static/js/787.0fef8e45.chunk.js": "/static/js/787.0fef8e45.chunk.js",
+    "index.html": "/index.html",
+    "main.ee59dfc2.css.map": "/static/css/main.ee59dfc2.css.map",
+    "main.d05767f4.js.map": "/static/js/main.d05767f4.js.map",
+    "787.0fef8e45.chunk.js.map": "/static/js/787.0fef8e45.chunk.js.map"
+  },
+  "entrypoints": [
+    "static/css/main.ee59dfc2.css",
+    "static/js/main.d05767f4.js"
+  ]
+}

view/build/index.html

@@ -0,0 +1 @@
+<!doctype html><html lang="en"><head><meta charset="utf-8"/><link rel="icon" href="/favicon.ico"/><link rel="preconnect" href="https://fonts.googleapis.com"><link rel="preconnect" href="https://fonts.gstatic.com" crossorigin><link href="https://fonts.googleapis.com/css2?family=Kanit:wght@500&display=swap" rel="stylesheet"><meta name="viewport" content="width=device-width,initial-scale=1"/><meta name="theme-color" content="#000000"/><meta name="description" content="Web site created using create-react-app"/><link rel="apple-touch-icon" href="/logo192.png"/><link rel="manifest" href="/manifest.json"/><title>React App</title><script defer="defer" src="/static/js/main.d05767f4.js"></script><link href="/static/css/main.ee59dfc2.css" rel="stylesheet"></head><body><noscript>You need to enable JavaScript to run this app.</noscript><div id="root"></div></body></html>

view/build/manifest.json

@@ -0,0 +1,25 @@
+{
+  "short_name": "React App",
+  "name": "Create React App Sample",
+  "icons": [
+    {
+      "src": "favicon.ico",
+      "sizes": "128x128 64x64 32x32 24x24 16x16",
+      "type": "image/x-icon"
+    },
+    {
+      "src": "logo192.png",
+      "type": "image/png",
+      "sizes": "192x192"
+    },
+    {
+      "src": "logo512.png",
+      "type": "image/png",
+      "sizes": "512x512"
+    }
+  ],
+  "start_url": ".",
+  "display": "standalone",
+  "theme_color": "#000000",
+  "background_color": "#ffffff"
+}

view/build/robots.txt

@@ -0,0 +1,3 @@
+# https://www.robotstxt.org/robotstxt.html
+User-agent: *
+Disallow: