#Nn models sets vlad update#
#Nn models sets vlad full#

medvram Enable model optimizations for sacrificing a little speed for low memory usage no-half-vae Do not switch the VAE model to 16-bit floats no-half Do not switch the model to 16-bit floats h, -help show this help message and exit This includes choices of cross-optimization and memory attention as well as configuration of system paths Until I make CSS cleaner, this is a bit ugly so I do not recommend using 3rd party themes for now Moved almost all command line params to settings configurable from UI To reset settings to defaults, delete config.json and restart server UI changes Quite a few defaults have been changed, but if you're have your config.json, it will remain as-is Yes, requirements-versions.txt is gone Settings changes More work to be done, but all essentials are in place Built-in performance profilerĮnable with -profile flag Simplified dependencies and updated core libraries New logger and new error and exception handlers This can be disabled with cmd flag -noupdateĪlternatively, you can run in quick mode if you're sure everything is up-to-date by adding cmd flag -quickītw, if you're wondering what takes most time during startup: its running installers for each extension - so having large number of extensions installed has a major performance impact on server startup timeġ1:05:33-591557 INFO Python 3.10.6 on Linuxġ1:05:33-593987 INFO Installing requirementsġ1:05:33-651731 INFO Version: 81b8294e Wed Apr 12 10:40:11 2023 -0400ġ1:05:33-945552 WARNING Latest available version: ġ1:05:35-695232 INFO Installing repositoriesġ1:05:35-700296 INFO Installing submodulesġ1:05:39-769383 INFO Built-in extensions: ġ1:05:49-397596 INFO Enabled extensions: ġ1:05:52-896265 INFO Installing requirementsġ1:05:52-899582 INFO Setting environment tuningġ1:05:52-900085 INFO Server Arguments: ġ1:05:56-195850 INFO Torch 2.0.0+cu118 CUDA 11.8 cuDNN 8700ġ1:05:56-197752 INFO GPU NVIDIA GeForce RTX 3060 VRAM 12288 Arch (8, 6) Cores 28

#Nn models sets vlad update#

Note: by default, launcher/setup will re-check dependencies and update extensions and submodules on each startup. If running manually, make sure to activate venv first (if used) If you're unsure which launcher to use, this is the one you wantĬan be used directly to start server in manually activated venv or to run it without venvĬan also be used directly to update repository or extensions Note: Server can run without virtual environment, but it is recommended to use it Platform specific wrapper scripts that starts launch.py in Python virtual environment This should be fully cross-platform, but I would really love to have additional contibutors and/or maintainers to join and help lead the effords on different platforms How WebUI is started is pretty much completely changed:

The goal is to actually integrate all best options for uniform out-of-the-box experience without the need to tweak anything manually. If you're interested, ping me!Īdditionally, I'd love area-experts for different platform to help with best practices:

#Nn models sets vlad full#

To avoid having this repo rely just on me, I'd love to have additional maintainers with full admin rights. This has been pending for a while, but finally uploaded some massive changes Ask for help