16 commits

Author	SHA1	Message	Date
Max	079f83ba45	Major Upgrade/Refactor ... - KoboldAI is no longer upstream maintained, so is now deprecated in nixified.ai - InvokeAI: v2.3.1.post2 -> v3.3.0post3 - textgen: init at v1.7 - treewide: update flake inputs including nixpkgs - treewide: add a bunch of new dependencies and upgrade old ones	2023-10-19 22:13:37 +01:00
Max	776fa40bb2	invokeai/nixos: remove AMD-specific workarounds	2023-04-12 18:29:25 +02:00
Max	44209b89f9	invokeai/nixos: use PrivateTmp	2023-04-12 18:29:25 +02:00
Max	781827ea87	invokeai: apply ROCm workarounds	2023-04-12 18:29:25 +02:00
Max	11ab758490	invokeai: add NixOS option for model precision	2023-03-13 20:59:58 +01:00
Max	6bfc491e80	invokeai: set HOME in systemd service	2023-03-13 20:59:58 +01:00
Max	46e6505660	invokeai: add a NixOS module for AMD-specifc quirks	2023-03-13 20:59:58 +01:00
Max	e110b678dc	invokeai: add NixOS modules	2023-03-13 20:59:58 +01:00
arcnmx	2f69fda039	flake: callPackage streamlit	2023-03-06 09:27:39 -08:00
arcnmx	11ea9ebc5d	flake: update	2023-03-06 08:20:53 -08:00
matthewcroughan	4394b44c93	invokeai: update to v2.3.1.post2 ... Co-authored-by: Max <max@privatevoid.net>	2023-03-03 03:38:31 +00:00
matthewcroughan	339a78b0e2	invokeai: support WSL via wrapper, not extra outputs ... This removes the need to have multiple output variations like invokeai-wsl-amd and instead just uses a simple bash wrapper to discover whether the program is being ran in the WSL	2023-02-25 22:19:40 +00:00
Max	70fd1f1831	invokeai: use callPackage	2023-02-20 22:04:48 +01:00
matthewcroughan	a91a1a6db1	invokeai: add wsl variants ... This sets the LD_LIBRARY_PATH to /usr/lib/wsl/lib so that invokeai can use a Windows Host's GPU drivers	2023-02-13 01:55:35 +00:00
Max	d6072e2599	projects/invokeai: add metadata	2023-02-12 01:11:15 +01:00
Max	ce390ad0ed	cleanup and modularize	2023-02-12 01:04:57 +01:00