Commit Graph

Select branches

Hide Pull Requests

main

Mono Color

• 1c5fe809c2 . Hacker 17082006 2023-02-17 17:59:16 +0700
• 1cf5b6125b Merge branch 'hide-triton-error-windows' of https://github.com/pythongosssss/ComfyUI comfyanonymous 2023-02-17 02:40:42 -0500
• 638f924546 Sort the list of input images. comfyanonymous 2023-02-16 21:01:46 -0500
• 6135a21ee8 Add a way to control controlnet strength. comfyanonymous 2023-02-16 18:08:01 -0500
• 0787299610
  Hide Triton warning on Windows pythongosssss 2023-02-16 18:19:26 +0000
• 56498d505a Create controlnet directory. comfyanonymous 2023-02-16 10:50:30 -0500
• 4efa67fa12 Add ControlNet support. comfyanonymous 2023-02-16 10:38:08 -0500
• bc69fb5245 Use inpaint models the proper way by using VAEEncodeForInpaint. comfyanonymous 2023-02-15 20:44:51 -0500
• 59bef84bc8 Add the config for SD2.x inpainting models. comfyanonymous 2023-02-15 17:52:34 -0500
• e87a8669b6 Add a LoadImageMask node to load one colour channel in an image as a mask. comfyanonymous 2023-02-15 17:39:42 -0500
• d75003001a Add a SetLatentNoiseMask node. comfyanonymous 2023-02-15 16:58:55 -0500
• cef2cc3cb0 Support for inpaint models. comfyanonymous 2023-02-15 16:38:20 -0500
• 07db00355f Add masks to samplers code for inpainting. comfyanonymous 2023-02-15 01:49:17 -0500
• e1a047971b Edit comment Hacker 17082006 2023-02-15 21:50:18 +0700
• 7da0147d8a Done custom_node import error fixing Hacker 17082006 2023-02-15 21:48:10 +0700
• b67b6920d1 Fix bug Hacker 17082006 2023-02-15 21:15:54 +0700
• 4b4eae7171 Merge branch 'master' of https://github.com/Fannovel16/ComfyUI Hacker 17082006 2023-02-14 22:20:53 +0700
• 493671574e Adding some suggests from the author Hacker 17082006 2023-02-14 22:20:30 +0700
• 162cc04247
  Merge branch 'comfyanonymous:master' into master Fannovel16 2023-02-14 12:52:06 +0700
• c1d5810020 Another small fix. comfyanonymous 2023-02-13 16:07:34 -0500
• 73251f92c8 Now it works in colab properly again. comfyanonymous 2023-02-13 16:04:26 -0500
• 93d84e3b9c Lower timeout and fix for colab. comfyanonymous 2023-02-13 15:52:34 -0500
• 68a1ac7746
  Remove catch that was causing linux ctrl+c issues pythongosssss 2023-02-13 20:12:12 +0000
• 8e51af6786
  Changed to serve index.html on root pythongosssss 2023-02-13 17:58:03 +0000
• 540bbbdb82
  Handle https when creating socket pythongosssss 2023-02-13 17:53:21 +0000
• c472377622
  Merge branch 'comfyanonymous:master' into master pythongosssss 2023-02-13 17:48:36 +0000
• e3451cea4f uni_pc now works with KSamplerAdvanced return_with_leftover_noise. comfyanonymous 2023-02-13 12:29:21 -0500
• 721f8510af Change the words a bit Hacker 17082006 2023-02-13 18:24:54 +0700
• f3c6d7adf2 . Hacker 17082006 2023-02-13 18:19:42 +0700
• 5ce4b9eca3 Add custom node feature Hacker 17082006 2023-02-13 18:17:40 +0700
• f616dcc81a
  Added handling of missing aiohttp module pythongosssss 2023-02-12 18:32:20 +0000
• 639d21e5f7
  Added aiohttp requirement pythongosssss 2023-02-12 18:31:26 +0000
• f558942877
  Merge branch 'comfyanonymous:master' into master pythongosssss 2023-02-12 18:30:53 +0000
• 3fd6b7027c Support dark mode in GUI. comfyanonymous 2023-02-12 13:29:34 -0500
• 05ad64d22c Add a feather option to the latent composite node. comfyanonymous 2023-02-12 13:01:52 -0500
• 9d1edfde1f
  Changed polling to use websockets pythongosssss 2023-02-12 15:54:22 +0000
• 5d14e9b959
  Changed HTTP Server + Added WebSockets pythongosssss 2023-02-12 15:53:48 +0000
• f542f248f1 Show the right amount of steps in the progress bar for uni_pc. comfyanonymous 2023-02-11 14:59:42 -0500
• f10b8948c3 768-v support for uni_pc sampler. comfyanonymous 2023-02-11 04:34:58 -0500
• ce0aeb109e Remove print. comfyanonymous 2023-02-11 03:41:40 -0500
• 5489d5af04 Add uni_pc sampler to KSampler* nodes. comfyanonymous 2023-02-11 03:18:27 -0500
• 1a4edd19cd Fix overflow issue with inplace softmax. comfyanonymous 2023-02-10 11:47:41 -0500
• 509c7dfc6d Use real softmax in split op to fix issue with some images. comfyanonymous 2023-02-10 02:46:50 -0500
• 7e1e193f39 Automatically enable lowvram mode if vram is less than 4GB. comfyanonymous 2023-02-10 00:47:56 -0500
• e9d3ac2ba0 Typo. comfyanonymous 2023-02-09 16:31:54 -0500
• 9ae4e1b0ff Add a list of features to the Readme. comfyanonymous 2023-02-09 16:27:52 -0500
• 324273fff2 Fix embedding not working when on new line. comfyanonymous 2023-02-09 14:12:02 -0500
• 1f6a467e92 Update ldm dir with latest upstream stable diffusion changes. comfyanonymous 2023-02-09 13:47:36 -0500
• 642516a3a6 create output dir if none is present BazettFraga 2023-02-09 18:32:23 +0100
• 773cdabfce Same thing but for the other places where it's used. comfyanonymous 2023-02-09 12:43:29 -0500
• df40d4f3bf torch.cuda.OutOfMemoryError is not present on older pytorch versions. comfyanonymous 2023-02-09 12:33:27 -0500
• 1d9ec62cfb Use absolute output directory path. comfyanonymous 2023-02-09 09:59:43 -0500
• 05d571fe7f Merge branch 'master' of https://github.com/bazettfraga/ComfyUI into merge_pr2 comfyanonymous 2023-02-09 00:44:38 -0500
• e8c499ddd4 Split optimization for VAE attention block. comfyanonymous 2023-02-08 21:54:44 -0500
• 5b4e312749 Use inplace operations for less OOM issues. comfyanonymous 2023-02-08 20:52:02 -0500
• e58887dfa7 forgot windows does double backslashes for paths due to its use as escape char. BazettFraga 2023-02-09 01:30:06 +0100
• 81082045c2 add recursive_search, swap relevant os.listdirs BazettFraga 2023-02-09 01:22:33 +0100
• 3fd87cbd21 Slightly smarter batching behaviour. comfyanonymous 2023-02-08 17:09:47 -0500
• bbdcf0b737 Use relative imports for k_diffusion. comfyanonymous 2023-02-08 16:51:19 -0500
• 3e22815a9a Fix k_diffusion not getting imported from the folder. comfyanonymous 2023-02-08 16:29:22 -0500
• 708138c77d Remove print. comfyanonymous 2023-02-08 14:51:18 -0500
• 047775615b Lower the chances of an OOM. comfyanonymous 2023-02-08 14:24:27 -0500
• 853e96ada3 Increase it/s by batching together some stuff sent to unet. comfyanonymous 2023-02-08 14:05:31 -0500
• c92633eaa2 Auto calculate amount of memory to use for --lowvram comfyanonymous 2023-02-08 11:42:37 -0500
• 534736b924 Add some low vram modes: --lowvram and --novram comfyanonymous 2023-02-08 11:37:10 -0500
• a84cd0d1ad Don't unload/reload model from CPU uselessly. comfyanonymous 2023-02-08 03:17:54 -0500
• e3e65947f2 Add a --help to main.py comfyanonymous 2023-02-07 22:12:56 -0500
• 1f18221e17 Add --port to set custom port. comfyanonymous 2023-02-07 21:57:17 -0500
• 6e40393b6b Fix delete sometimes not properly refreshing queue state. comfyanonymous 2023-02-07 00:07:31 -0500
• d71d0c88e5 Add some simple queue management to the GUI. comfyanonymous 2023-02-06 23:40:38 -0500
• b1a7c9ebf6 Embeddings/textual inversion support for SD2.x comfyanonymous 2023-02-05 15:49:03 -0500
• 1de5aa6a59 Add a CLIPLoader node to load standalone clip weights. comfyanonymous 2023-02-05 15:20:18 -0500
• 56d802e1f3 Use transformers CLIP instead of open_clip for SD2.x comfyanonymous 2023-02-05 14:36:28 -0500
• bf9ccffb17 Small fix for SD2.x loras. comfyanonymous 2023-02-05 11:38:25 -0500
• 678105fade SD2.x CLIP support for Loras. comfyanonymous 2023-02-05 01:54:09 -0500
• 3f3d77a324 Fix image node always executing instead of only when the image changed. comfyanonymous 2023-02-04 16:08:29 -0500
• 4225d1cb9f Add a basic ImageScale node. comfyanonymous 2023-02-04 15:53:29 -0500
• bff0e11941 Add a LatentCrop node. comfyanonymous 2023-02-04 15:21:46 -0500
• 43c795f462 Add a --listen argument to listen on 0.0.0.0 comfyanonymous 2023-02-04 12:01:53 -0500
• 41a7532c15 A bit bigger. comfyanonymous 2023-02-03 13:56:00 -0500
• 7bc3f91bd6 Add some instructions how to use the venv from another SD install. comfyanonymous 2023-02-03 13:54:45 -0500
• 149a4de3f2 Fix potential issue if exception happens when patching model. comfyanonymous 2023-02-03 03:55:50 -0500
• ef90e9c376 Add a LoraLoader node to apply loras to models and clip. comfyanonymous 2023-02-03 02:06:34 -0500
• 96664f5d5e Web interface bug fix for multiple inputs from the same node. comfyanonymous 2023-02-03 00:39:28 -0500
• 1d84a44b08 Fix some small annoyances with the UI. comfyanonymous 2023-02-02 14:09:14 -0500
• e65a20e62a Add a button to queue prompts to the front of the queue. comfyanonymous 2023-02-01 22:34:59 -0500
• 4b08314257 Add more features to the backend queue code. comfyanonymous 2023-02-01 22:33:10 -0500
• 9d611a90e8 Small web interface fixes. comfyanonymous 2023-01-31 03:37:34 -0500
• fef41d0a72 Add LatentComposite node. comfyanonymous 2023-01-31 03:35:03 -0500
• 3fa009f4cc Add a LatentFlip node. comfyanonymous 2023-01-31 03:28:38 -0500
• 69df7eba94 Add KSamplerAdvanced node. comfyanonymous 2023-01-31 03:09:38 -0500
• f8f165e2c3 Add a LatentRotate node. comfyanonymous 2023-01-31 02:28:07 -0500
• 1daccf3678 Run softmax in place if it OOMs. comfyanonymous 2023-01-30 19:55:01 -0500
• 0d8ad93852 Add link to examples github page. comfyanonymous 2023-01-30 01:09:35 -0500
• f73e57d881 Add support for textual inversion embedding for SD1.x CLIP. comfyanonymous 2023-01-29 18:46:44 -0500
• 702ac43d0c Readme formatting. comfyanonymous 2023-01-29 13:23:57 -0500
• da6f56235b Add section to readme explaining how to get better speeds. comfyanonymous 2023-01-29 13:15:03 -0500
• 3661e10648 Add a command line option to disable upcasting in some cross attention ops. comfyanonymous 2023-01-29 13:12:22 -0500
• 50db297cf6 Try to fix OOM issues with cards that have less vram than mine. comfyanonymous 2023-01-29 00:50:46 -0500
• 36ec5690a6 Add some more model configs including some to use SD1 models in fp16. comfyanonymous 2023-01-28 23:23:49 -0500