Deployment Package Author: Akiba aaaki
Disclaimer:
This installation package and launcher are provided for free without any profit purpose
Hello everyone, I am rain or shine. As we all know, StableDiffusion is a very powerful AI drawing tool. Friends who need to know more about StableDiffusion can check out my previous article: Midjourney VS StableDiffusion, two of the most popular AI drawing tools recently .
So starting from this issue, I will continue to update the Stable Diffusion research series of articles, taking you from entry to mastery, and playing with Stable Diffusion.
Today I bring you Stable Diffusion Research (1) sd integration package v4.2 installation instructions.
The installation version brought here is the integrated package sd-webui-aki-v4.2 version .
1. Detailed description of the integration package
1. Contents of the integration package upgrade:
-
torch2, xformers0.0.17, cudnn8.8 can be opened at full speed without any operation (including 40 series graphics card)
-
Upgrade various other dependent versions
-
Preset the model of Tagger (image inversion keyword)
-
Preset ControlNet, MultiDiffusion plug-in optimized some other settings
2. Integration package statement:
- The integration package is based on the open source project stable defusing web ui to make ai paintings. There has never been any official version, and it will not be any different from the so-called self-deploying official version.
- Generally speaking, if something goes wrong, it must be a problem with your own use.
- The integration package only packs the necessary environment for operation, such as python and git. And the model is prefabricated, and some commonly used plug-ins are added. All environments are in this package, which can be deleted as needed.
- Environments are virtual independently and do not create any conflicts. In theory, it is more stable than the one deployed by yourself. Because there are really too many problems in my own deployment.
In summary, it should be the version with the best speed at present.
3. Suitable for the crowd:
-
Zero-based beginners who have never used AI to paint
-
People who broke the original integrated package by themselves
-
people who don't want to use their brains
4. Configuration requirements required for operation:
-
The operating system requires windows10 or later
-
The system cpu is not mandatory
-
Memory recommended 8g or more
-
The graphics card must be a NAIDIA discrete graphics card with a minimum of 4g of video memory
-
Solid-state hard disk, the best space is more than 200g, to improve the model loading speed
5. The integration package includes:
Optional ControlNet model : You don’t need to download it, it’s just provided here, download it on demand, and refer to my other tutorial for the installation method: AI Painting Stable Diffusion Research (2) Introduction and installation of sd model ControlNet1.1 .
Launcher operation dependency -donet-6.0.11.exe: Launcher operation dependency, if you have not used the launcher before, you need to install it.
sd-webui-aki-v4.zip: integration package
As shown in the picture:
2. Installation steps
1. Unzip sd-webui-aki-v4.zip , needless to say.
2. If you haven’t used the launcher before, you need to install the dependencies of the launcher first: the launcher’s operation depends on -donet-6.0.11.exe , and the installation can be completed in the next step.
3. After the installation is complete, enter the decompressed integrated package directory: sd-webui-aki-v4, and double-click the A launcher.exe to run it.
After startup as follows:
4. Click the "One-key start" button to start the Stable Diffusion console and operation interface.
As shown in the picture:
3. Subsequent version update
Subsequent version updates and plug-in updates are all done in the launcher.
1. Click version management on the left
2. Upgrade the main body
Switch to the Ontology tab, and click the "One-click Update" button in the upper right corner.
As shown in the picture:
3. Upgrade the extension plug-in
Switch to the extension tab and click the "One-click update" button in the upper right corner.
So far, the installation instruction of sd integration package v4.2 ends.
The next article will bring you the installation and use of the model ControlNet1.1, so stay tuned.