Commit Graph

2822 Commits

Author SHA1 Message Date
comfyanonymous 1122df1a20 Increase range of lora strengths. 2023-09-20 17:58:54 -04:00
comfyanonymous 1cdfb3dba4 Only do the cast on the device if the device supports it. 2023-09-20 17:52:41 -04:00
comfyanonymous b92a86d737 Update litegraph to upstream. 2023-09-20 13:24:08 -04:00
comfyanonymous f895260e5e Merge branch 'escape-glob' of https://github.com/seanlynch/ComfyUI 2023-09-19 13:13:40 -04:00
comfyanonymous 7c9a92f552 Don't depend on torchvision. 2023-09-19 13:12:47 -04:00
Sean Lynch 8321592408 Escape paths when passing them to globs
Try to prevent JS search from breaking on pathnames with square
brackets.
2023-09-19 08:18:29 -04:00
MoonRide303 2b6b178173 Added support for lanczos scaling 2023-09-19 10:40:38 +02:00
comfyanonymous 6d3dee9d16 Clean up #1541. 2023-09-18 23:33:52 -04:00
comfyanonymous f32463936d Unhardcode sampler and scheduler list in test. 2023-09-18 23:24:14 -04:00
City 7c93afd2cd
Manual float precision, toggle for old behavior (#1541)
* Add toggle for float rounding

* Add manual precision override
2023-09-18 23:20:00 -04:00
enzymezoo-code 26cd8405dd
Ci quality workflows (#1423)
* Add inference tests

* Clean up

* Rename test graph file

* Add readme for tests

* Separate server fixture

* test file name change

* Assert images are generated

* Clean up comments

* Add __init__.py so tests can run with command line `pytest`

* Fix command line args for pytest

* Loop all samplers/schedulers in test_inference.py

* Ci quality workflows compare (#1)

* Add image comparison tests

* Comparison tests do not pass with empty metadata

* Ensure tests are run in correct order

* Save image files  with test name

* Update tests readme

* Reduce step counts in tests to ~halve runtime

* Ci quality workflows build (#2)

* Add build test github workflow
2023-09-18 23:18:06 -04:00
comfyanonymous b92bf8196e Do lora cast on GPU instead of CPU for higher performance. 2023-09-18 23:04:49 -04:00
comfyanonymous 0109431626 Lower the minimum resolution of EmptyLatentImage. 2023-09-18 16:20:03 -04:00
comfyanonymous db63aa7e53 Nodes can now control the rounding in the UI. 2023-09-17 12:49:06 -04:00
comfyanonymous 321c5fa295 Enable pytorch attention by default on xpu. 2023-09-17 04:09:19 -04:00
comfyanonymous 0665749b1a Move ModelSubtract and ModelAdd to advanced/model_merging 2023-09-17 02:10:06 -04:00
comfyanonymous d6d9b83447 Merge branch 'fix/preview_ratio' of https://github.com/ltdrdata/ComfyUI 2023-09-16 15:43:42 -04:00
comfyanonymous 61b1f67734 Support models without previews. 2023-09-16 12:59:54 -04:00
Dr.Lt.Data 4d5e057bb2 fix indent 2023-09-16 20:37:42 +09:00
Dr.Lt.Data 69680fede7 fix: thumbnail ratio fix for mixed ratio images 2023-09-16 20:36:00 +09:00
comfyanonymous 43d4935a1d Add cond_or_uncond array to transformer_options so hooks can check what is
cond and what is uncond.
2023-09-15 22:21:14 -04:00
comfyanonymous 415abb275f Add DDPM sampler. 2023-09-15 19:22:47 -04:00
comfyanonymous 099226015e Merge branch 'Fix-structuredClone-error-with-early-chrome-version-browser' of https://github.com/KarryCharon/ComfyUI 2023-09-15 15:48:22 -04:00
comfyanonymous 94e4fe39d8 This isn't used anywhere. 2023-09-15 12:03:03 -04:00
karrycharon 076f3e6310 fix structuredClone undefined error; 2023-09-15 16:37:58 +08:00
comfyanonymous 44361f6344 Support for text encoder models that need attention_mask. 2023-09-15 02:02:05 -04:00
comfyanonymous 0d8f376446 Set last layer on SD2.x models uses the proper indexes now.
Before I had made the last layer the penultimate layer because some
checkpoints don't have them but it's not consistent with the others models.

TLDR: for SD2.x models only: CLIPSetLastLayer -1 is now -2.
2023-09-14 20:28:22 -04:00
comfyanonymous 0966d3ce82 Don't run text encoders on xpu because there are issues. 2023-09-14 12:16:07 -04:00
pythongosssss 0e4395a8a3 Allow pasting nodes with connections in firefox 2023-09-13 18:42:44 +01:00
comfyanonymous 3039b08eb1 Only parse command line args when main.py is called. 2023-09-13 11:38:20 -04:00
comfyanonymous 30de95e4b4 Add some nodes to subtract and add model weights. 2023-09-13 01:10:31 -04:00
comfyanonymous 0b829fe35b .gitignore refactor. 2023-09-12 18:44:05 -04:00
comfyanonymous ed58730658 Don't leave very large hidden states in the clip vision output. 2023-09-12 15:09:10 -04:00
comfyanonymous fb3b728203 Fix issue where autocast fp32 CLIP gave different results from regular. 2023-09-11 21:49:56 -04:00
comfyanonymous 7d401ed1d0 Add ldm format support to UNETLoader. 2023-09-11 16:36:50 -04:00
comfyanonymous 9562a6b49e Fix a few clipboard issues. 2023-09-10 11:19:31 -04:00
comfyanonymous d4b2bc0964 Merge branch 'master' of https://github.com/miabrahams/ComfyUI 2023-09-10 10:15:02 -04:00
comfyanonymous 122fd5d37f Merge branch 'add-defaultInput' of https://github.com/chrisgoringe/ComfyUI 2023-09-10 03:18:05 -04:00
comfyanonymous 7df822212f Allow checkpoints with .pt and .bin extensions. 2023-09-10 02:36:04 -04:00
comfyanonymous 07691e80c3 Does it make sense to allow configuring the round and precision? 2023-09-09 03:15:31 -04:00
comfyanonymous 5c8b7ea03c Merge branch 'round-float-widgets' of https://github.com/chrisgoringe/ComfyUI 2023-09-09 03:07:57 -04:00
Chris 7372255e49 Specify the precision and rounding based on step 2023-09-09 15:21:38 +10:00
Michael Poutre cc2fa311dd
fix(server): Disable access logs 2023-09-08 21:11:53 -07:00
comfyanonymous e85be36bd2 Add a penultimate_hidden_states to the clip vision output. 2023-09-08 14:06:58 -04:00
comfyanonymous 10de64af7f Google doesn't want people to use ComfyUI on colab anymore. 2023-09-08 14:02:03 -04:00
Michael Abrahams 264867bf87 Clear clipboard on copy 2023-09-08 12:42:13 -04:00
comfyanonymous 1e6b67101c Support diffusers format t2i adapters. 2023-09-08 11:36:51 -04:00
Chris 3ebe6b539a round float widgets (by default to 0.001) 2023-09-08 20:40:27 +10:00
MoonRide303 ff962098fd Fixed Load Image preview not displaying some files (issue #1158) 2023-09-08 08:43:17 +02:00
Chris 0782ac2a96 defaultInput 2023-09-08 14:53:59 +10:00