Differences

This shows you the differences between two versions of the page.

--- ai:stable-diffusion [2024/04/27 15:23] – [Linux Docker install] Wulf Rajek
+++ ai:stable-diffusion [2024/07/11 13:00] (current) – Wulf Rajek
@@ Line 48: / Line 48: @@
   - Download and run installers for [[https://www.python.org/ftp/python/3.10.6/python-3.10.6-amd64.exe|Python 3.10.6]] (ticking Add to PATH) and [[https://git-scm.com/download/win|git]]
-  - Clone the fork as AMD is not officially supported\\ Paste into command prompt in a directory where it should live\\ <code>git clone https://github.com/lshqqytiger/stable-diffusion-webui-directml && cd stable-diffusion-webui-directml && git submodule init && git submodule update
+  - Clone the fork as AMD is not officially supported\\ Paste into command prompt in a directory where it should live\\ <code>git clone https://github.com/lshqqytiger/stable-diffusion-webui-directml
+cd stable-diffusion-webui-directml
+git submodule init
+git submodule update
+venv\Scripts\python.exe -m pip install --upgrade pip
+venv\Scripts\pip.exe pip install -r requirements.txt
+venv\Scripts\pip.exe install torch-directml
 </code>
-  - If you have only 4-6gb vram, try adding these flags to webui-user.bat like so:\\ <code>COMMANDLINE_ARGS=--opt-sub-quad-attention --lowvram --disable-nan-check</code>
+  - Right-click webui-user.bat and adjust the command line args to\\ <code>COMMANDLINE_ARGS=--use-directml --skip-torch-cuda-test</code>
-  - Double-click webui-user.bat in that directory
+  - If you have only 4-6gb vram, try adding these additional flags to webui-user.bat's command line argslike so:\\ <code>--opt-sub-quad-attention --lowvram --disable-nan-check</code>
+  - Double-click webui-user.bat in that directory, this will download ~2.7GB torch and other python modules, followed by the 4GB Stable-Diffusion 1.5 pruned model.
   - If it looks like it is stuck when installing or running, press enter in the terminal and it should continue.
-  - point browser to [[http://localhost:7860/]]
+  - once done, a browser opens to [[http://localhost:7860/]]
-Note: This is only a GUI. AI Models need to be downloaded to generate anything.
+Check/Test if GPU is used in Task Manager's GPU tab. Dedicated GPU Memory should show large usage for the loaded Model and when generating a test image in the web interface, the GPU usage should ramp up like here:\\
+{{:ai:pasted:20240504-020430.png}}
+===== Performance =====
+Approx times to generate 512x512 default settings images:
+^CPU/GPU^Model^Time^
+|CPU|Intel 8th gen i5 (Intel NUC)|5-15 minutes|
+|GPU|AMD Radeon 780M|1.5 minutes|
+|GPU|AMD Radeon 7900XTX|21.7 seconds|
+|GPU|Nvidia RTX 4080|1.2 seconds|
+===== Gallery Addon =====
+As an extension for SD-webui:
+Open the Extensions tab in SD-webui.
+  - Select the Install from URL option.
+  - Enter https://github.com/zanllp/sd-webui-infinite-image-browsing
+  - Click on the Install button.
+  - Wait for the installation to complete and click on Apply and restart UI.
+{{:ai:pasted:20240518-142918.png?500}}
+===== Cozy Nest UI Addon =====
+As an extension for SD-webui:
+Open the Extensions tab in SD-webui.
+  - Select the Install from URL option.
+  - Enter https://github.com/Nevysha/Cozy-Nest
+  - Click on the Install button.
+  - Wait for the installation to complete and click on Apply and restart UI.
+{{:ai:pasted:20240518-142848.png?500}}
+===== general advice =====
 [[https://www.reddit.com/r/StableDiffusion/comments/x41n87/how_to_get_images_that_dont_suck_a/|Some general Stable Diffusion prompt advice]]
+[[https://machinelearningmastery.com/more-prompting-techniques-for-stable-diffusion/|More promptng techniques]]
+[[https://learnopencv.com/sdxl-inpainting/|Inpainting explanation]]
 ====== Models / Settings ======
+LoRA models/styles/tools as well as base models can be searched and downloaded from [[https://civitai.com/models|CivitAI]]
 ===== Realistic Vision + Lora =====
@@ Line 99: / Line 143: @@
 download [[https://huggingface.co/stabilityai/sdxl-turbo/resolve/main/sd_xl_turbo_1.0_fp16.safetensors|SDXL Turbo]] from https://huggingface.co/stabilityai/sdxl-turbo and copy into ''Models/Stable-diffusion'' directory.
+===== Text LoRA =====
-===== Test image prompts =====
+https://civitai.com/models/176555?modelVersionId=214296
+Other Text LoRAs:
+https://civitai.com/tag/text
+===== Image Style/Prompt libraries =====
+[[https://lexica.art/|Lexica]]
+[[https://prompthero.com/|PromptHero]]
+===== Some test image prompts =====
 <code>
@@ Line 130: / Line 184: @@
 {{:ai:pasted:20240427-095226.png}}
+====== Update docker ======
+https://github.com/AbdBarho/stable-diffusion-webui-docker/issues/683
+If you want a more up to date version of A1111 here are the steps:
+<code>
+docker exec -it webui-docker-auto-1 /bin/bash
+git pull
+pip install -r requirements.txt
+#if needed
+cd repositories
+git clone https://github.com/AUTOMATIC1111/stable-diffusion-webui-assets.git
+exit
+docker stop webui-docker-auto-1
+docker start webui-docker-auto-1
+</code>
+If it fails for any reason just rebuild the container using the standard process of docker compose --profile auto up --build