Commit Graph

7075 Commits

Author SHA1 Message Date
28208e6f49 fix: Fix VAE Precision not working for SDXL Canvas Modes 2023-08-13 04:09:51 +12:00
c33acf951e feat: Make Refiner work with Canvas 2023-08-13 03:53:40 +12:00
500cd552bc feat: Make SDXL work across the board + Custom VAE Support
Also a major cleanup pass to the SDXL graphs to ensure there's no ID overlap
2023-08-13 01:45:03 +12:00
55d27f71a3 feat: Give each graph its own unique id 2023-08-13 00:51:10 +12:00
746c7c59ff fix: remove extra node for canvas output catch 2023-08-12 22:39:30 +12:00
ad96c41156 feat: Add Canvas Output node to all Canvas Graphs 2023-08-12 22:04:43 +12:00
27bd127fb0 fix: Do not add anything but final output to staging area 2023-08-12 21:10:30 +12:00
f296e5c41e wip: Remove MaskBlur / Adjust color correction 2023-08-12 20:54:30 +12:00
a67d8376c7 fix missed spot for autoAddBoardId none 2023-08-12 18:07:01 +10:00
9f6221fe8c Merge branch 'main' into feat/refactor_generation_backend 2023-08-12 18:37:47 +12:00
7587b54787 chore: Cleanup, comment and organize Node Graphs
Before it gets too chaotic
2023-08-12 17:17:46 +12:00
7254ffc3e7 chore: Split Inpaint and Outpaint Graphs 2023-08-12 16:30:20 +12:00
6034fa12de feat: Add Mask Blur node 2023-08-12 16:20:58 +12:00
ce3675fc14 Apply denoising_start/end according on timestep value 2023-08-12 03:19:49 +03:00
8acd7eeca5 feat: Disable clip skip for SDXL Canvas 2023-08-12 08:18:30 +12:00
7293a6036a feat(wip): Add SDXL To Canvas 2023-08-12 08:16:05 +12:00
0b11f309ca instead of crashing when a corrupted model is detected, warn and move on 2023-08-11 15:05:14 -04:00
6a8eb392b2 Add support for loading SDXL LoRA weights in diffusers format. 2023-08-11 14:40:22 -04:00
f343ab0302 wip: Port Outpainting to new backend 2023-08-12 06:15:59 +12:00
824ca92760 fix maximum python version instructions 2023-08-11 13:49:39 -04:00
d7d6298ec0 feat: Add Infill Method support 2023-08-12 05:32:11 +12:00
58a48bf197 fix: LoRA list name sorting 2023-08-12 04:47:15 +12:00
5629d8fa37 fix; Key issue in Lora List 2023-08-12 04:43:40 +12:00
1affb7f647 feat: Add Paste / Mask Blur / Color Correction to Inpainting
Seam options are now removed. They are replaced by two options --Mask Blur and Mask Blur Method .. which control the softness of the mask that is being painted.
2023-08-12 03:28:19 +12:00
69a9dc7b36 wip: Add initial Inpaint Graph 2023-08-12 02:42:13 +12:00
f3ae52ff97 Fix error at high denoising_start, fix unipc(cpu_only) 2023-08-11 15:46:16 +03:00
7479f9cc02 feat: Update LinearUI to use new backend (except Inpaint) 2023-08-11 22:22:01 +12:00
87ce4ab27c fix: Update default_graph to use new DenoiseLatents 2023-08-11 22:21:13 +12:00
7c0023ad9e feat: Remove TextToLatents / Rename Latents To Latents -> DenoiseLatents 2023-08-11 22:20:37 +12:00
231e665675 Merge branch 'main' into feat/refactor_generation_backend 2023-08-11 20:53:38 +12:00
80fd4c2176 undo lint changes 2023-08-11 14:26:09 +10:00
3b6e425e17 fix error detail in toast 2023-08-11 14:26:09 +10:00
50415450d8 invalidate board total when images deleted, only run date range logic if board has less than 20 images 2023-08-11 14:26:09 +10:00
06296896a9 Update invokeai version v3.0.2 2023-08-10 22:23:41 -04:00
a7399aca0c Add new JS files for 3.0.2 build 2023-08-10 22:23:41 -04:00
d1ea8b1e98 Two changes to command-line scripts (#4235)
During install testing I discovered two small problems in the
command-line scripts. These are fixed.

## What type of PR is this? (check all applicable)

- [X Bug Fix

## Have you discussed this change with the InvokeAI team?
- [X] Yes
- 
      
## Have you updated all relevant documentation?
- [X] Yes


## Description

- installer - use correct entry point for invokeai-configure
- model merge script - prevent error when `--root` not provided
2023-08-10 21:11:45 -04:00
f851ad7ba0 Two changes to command-line scripts
- installer - use correct entry point for invokeai-configure
- model merge script - prevent error when `--root` not provided
2023-08-10 20:59:22 -04:00
591838a84b Add support for LyCORIS IA3 format (#4234)
## What type of PR is this? (check all applicable)

- [ ] Refactor
- [x] Feature
- [ ] Bug Fix
- [ ] Optimization
- [ ] Documentation Update
- [ ] Community Node Submission


## Have you discussed this change with the InvokeAI team?
- [ ] Yes
- [x] No

      
## Have you updated all relevant documentation?
- [ ] Yes
- [x] No


## Description
Add support for LyCORIS IA3 format

## Related Tickets & Documents
- Closes #4229 

## Added/updated tests?

- [ ] Yes
- [x] No
2023-08-11 03:30:35 +03:00
c0c2ab3dcf Format by black 2023-08-11 03:20:56 +03:00
56023bc725 Add support for LyCORIS IA3 format 2023-08-11 02:08:08 +03:00
2ef6a8995b Temporary force set vae to same precision as unet 2023-08-10 18:01:58 -04:00
d0fee93aac round slider values to nice numbers 2023-08-10 18:00:45 -04:00
1bfe9835cf clip cache settings to permissible values; remove redundant imports in install __init__ file 2023-08-10 18:00:45 -04:00
8e7eae6cc7 Probe LoRAs that do not have the text encoder (#4181)
## What type of PR is this? (check all applicable)

- [X] Bug Fix

## Have you discussed this change with the InvokeAI team?
- [X] No - minor fix

      
## Have you updated all relevant documentation?
- [X] Yes

## Description

It turns out that some LoRAs do not have the text encoder model, and
this was causing the code that distinguishes the model base type during
model import to reject them as having an unknown base model. This PR
enables detection of these cases.
2023-08-10 17:50:20 -04:00
f6522c8971 Merge branch 'main' into fix/detect-more-loras 2023-08-10 17:33:16 -04:00
a969707e45 prevent vae: '' from crashing model 2023-08-10 17:33:04 -04:00
6c8e898f09 Update scripts/verify_checkpoint_template.py
Co-authored-by: Eugene Brodsky <ebr@users.noreply.github.com>
2023-08-10 16:00:33 -04:00
7bad9bcf53 update dependencies and docs to cu118 2023-08-10 15:19:12 -04:00
d42b45116f fix(ui): fix lora sort (#4222)
## What type of PR is this? (check all applicable)

- [ ] Refactor
- [ ] Feature
- [s] Bug Fix
- [ ] Optimization
- [ ] Documentation Update
- [ ] Community Node Submission


## Have you discussed this change with the InvokeAI team?
- [x] Yes
- [ ] No, because:

      

## Description

was sorting with disabled at top of list instead of bottom

fixes #4217

## Related Tickets & Documents

<!--
For pull requests that relate or close an issue, please include them
below. 

For example having the text: "closes #1234" would connect the current
pull
request to issue 1234.  And when we merge the pull request, Github will
automatically close the issue.
-->

- Related Issue #
- Closes #4217

## QA Instructions, Screenshots, Recordings

<!-- 
Please provide steps on how to test changes, any hardware or 
software specifications as well as any other pertinent information. 
-->

![image](https://github.com/invoke-ai/InvokeAI/assets/4822129/dd895b86-05de-4303-8674-9b181037abaa)
2023-08-10 21:04:28 +12:00
d4812bbc8d Merge branch 'main' into fix/ui/fix-lora-sort 2023-08-10 19:00:26 +10:00