mlapointe
/
ComfyUI-Repo
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
2,552 Commits
1 Branch
0 Tags
13 MiB
main
Branches Tags
${ item.name }
Create tag ${ searchTerm }
Create branch ${ searchTerm }
from 'cd5017c1c9'
${ noResults }
Commit Graph

1014 Commits (cd5017c1c9b3a7b0fec892e80290a32616bbff38)

Author SHA1 Message Date
comfyanonymous cb7c4b4be3 Workaround for lora OOM on lowvram mode. 7 months ago
comfyanonymous 1208863eca Fix "Comfy" lora keys. 
They are in this format now:
diffusion_model.full.model.key.name.lora_up.weight
 7 months ago
comfyanonymous e1c528196e Fix bundled embed. 7 months ago
comfyanonymous 17030fd4c0 Support for "Comfy" lora format. 
The keys are just: model.full.model.key.name.lora_up.weight

It is supported by all comfyui supported models.

Now people can just convert loras to this format instead of having to ask
for me to implement them.
 7 months ago
comfyanonymous c19dcd362f Controlnet code refactor. 7 months ago
comfyanonymous 1c08bf35b4 Support format for embeddings bundled in loras. 7 months ago
comfyanonymous b334605a66 Fix OOMs happening in some cases. 
A cloned model patcher sometimes reported a model was loaded on a device
when it wasn't.
 7 months ago
comfyanonymous c14ac98fed Unload models and load them back in lowvram mode no free vram. 7 months ago
comfyanonymous 2d75df45e6 Flux tweak memory usage. 7 months ago
comfyanonymous 8edbcf5209 Improve performance on some lowend GPUs. 7 months ago
a-One-Fan a178e25912
Fix Flux FP64 math on XPU (#4210) 7 months ago
comfyanonymous 78e133d041 Support simple diffusers Flux loras. 7 months ago
Silver 7afa985fba
Correct spelling 'token_weight_pars_t5' to 'token_weight_pairs_t5' (#4200) 7 months ago
comfyanonymous 3b71f84b50 ONNX tracing fixes. 7 months ago
comfyanonymous 0a6b008117 Fix issue with some custom nodes. 7 months ago
comfyanonymous f7a5107784 Fix crash. 7 months ago
comfyanonymous 91be9c2867 Tweak lowvram memory formula. 7 months ago
comfyanonymous 03c5018c98 Lower lowvram memory to 1/3 of free memory. 7 months ago
comfyanonymous 2ba5cc8b86 Fix some issues. 7 months ago
comfyanonymous 1e68002b87 Cap lowvram to half of free memory. 7 months ago
comfyanonymous ba9095e5bd Automatically use fp8 for diffusion model weights if: 
Checkpoint contains weights in fp8.

There isn't enough memory to load the diffusion model in GPU vram.
 7 months ago
comfyanonymous f123328b82 Load T5 in fp8 if it's in fp8 in the Flux checkpoint. 7 months ago
comfyanonymous 63a7e8edba More aggressive batch splitting. 7 months ago
comfyanonymous ea03c9dcd2 Better per model memory usage estimations. 7 months ago
comfyanonymous 3a9ee995cf Tweak regular SD memory formula. 7 months ago
comfyanonymous 47da42d928 Better Flux vram estimation. 7 months ago
Alexander Brown ce9ac2fe05
Fix clip_g/clip_l mixup (#4168) 7 months ago
comfyanonymous e638f2858a Hack to make all resolutions work on Flux models. 7 months ago
comfyanonymous d420bc792a Tweak the memory usage formulas for Flux and SD. 7 months ago
comfyanonymous d965474aaa Make ComfyUI split batches a higher priority than weight offload. 7 months ago
comfyanonymous 1c61361fd2 Fast preview support for Flux. 7 months ago
comfyanonymous a6decf1e62 Fix bfloat16 potentially not being enabled on mps. 7 months ago
comfyanonymous 48eb1399c0 Try to fix mac issue. 7 months ago
comfyanonymous d7430a1651 Add a way to load the diffusion model in fp8 with UNETLoader node. 7 months ago
comfyanonymous f2b80f95d2 Better Mac support on flux model. 7 months ago
comfyanonymous 1aa9cf3292 Make lowvram more aggressive on low memory machines. 7 months ago
comfyanonymous eb96c3bd82 Fix .sft file loading (they are safetensors files). 7 months ago
comfyanonymous 5f98de7697 Load flux t5 in fp8 if weights are in fp8. 7 months ago
comfyanonymous 8d34211a7a Fix old python versions no longer working. 7 months ago
comfyanonymous 1589b58d3e Basic Flux Schnell and Flux Dev model implementation. 7 months ago
comfyanonymous 7ad574bffd Mac supports bf16 just make sure you are using the latest pytorch. 7 months ago
comfyanonymous e2382b6adb Make lowvram less aggressive when there are large amounts of free memory. 7 months ago
comfyanonymous c24f897352 Fix to get fp8 working on T5 base. 7 months ago
comfyanonymous a5991a7aa6 Fix hunyuan dit text encoder weights always being in fp32. 7 months ago
comfyanonymous 2c038ccef0 Lower CLIP memory usage by a bit. 7 months ago
comfyanonymous b85216a3c0 Lower T5 memory usage by a few hundred MB. 7 months ago
comfyanonymous 82cae45d44 Fix potential issue with non clip text embeddings. 7 months ago
comfyanonymous 25853d0be8 Use common function for casting weights to input. 7 months ago
comfyanonymous 79040635da Remove unnecessary code. 7 months ago
comfyanonymous 66d35c07ce Improve artifacts on hydit, auraflow and SD3 on specific resolutions. 
This breaks seeds for resolutions that are not a multiple of 16 in pixel
resolution by using circular padding instead of reflection padding but
should lower the amount of artifacts when doing img2img at those
resolutions.
 7 months ago
comfyanonymous 4ba7fa0244 Refactor: Move sd2_clip.py to text_encoders folder. 7 months ago
comfyanonymous cf4418b806 Don't treat Bert model like CLIP. 
Bert can accept up to 512 tokens so any prompt with more than 77 should
just be passed to it as is instead of splitting it up like CLIP.
 7 months ago
comfyanonymous 8328a2d8cd Let hunyuan dit work with all prompt lengths. 7 months ago
comfyanonymous afe732bef9 Hunyuan dit can now accept longer prompts. 7 months ago
comfyanonymous a9ac56fc0d Own BertModel implementation that works with lowvram. 7 months ago
comfyanonymous 25b51b1a8b Hunyuan DiT lora support. 7 months ago
comfyanonymous a5f4292f9f
Basic hunyuan dit implementation. (#4102) 
* Let tokenizers return weights to be stored in the saved checkpoint.

* Basic hunyuan dit implementation.

* Fix some resolutions not working.

* Support hydit checkpoint save.

* Init with right dtype.

* Switch to optimized attention in pooler.

* Fix black images on hunyuan dit.
 7 months ago
comfyanonymous f87810cd3e Let tokenizers return weights to be stored in the saved checkpoint. 7 months ago
comfyanonymous 10c919f4c7 Make it possible to load tokenizer data from checkpoints. 7 months ago
comfyanonymous 10b43ceea5 Remove duplicate code. 7 months ago
comfyanonymous 0a4c49c57c Support MT5. 7 months ago
comfyanonymous 88ed893034 Allow SPieceTokenizer to load model from a byte string. 7 months ago
comfyanonymous 334ba48cea More generic unet prefix detection code. 7 months ago
comfyanonymous 14764aa2e2 Rename LLAMATokenizer to SPieceTokenizer. 7 months ago
comfyanonymous b2c995f623 "auto" type is only relevant to the SetUnionControlNetType node. 7 months ago
Chenlei Hu 4151fbfa8a
Add error message on union controlnet (#4081) 7 months ago
comfyanonymous 95fa9545f1 Only append zero to noise schedule if last sigma isn't zero. 7 months ago
comfyanonymous 6ab8cad22e Implement beta sampling scheduler. 
It is based on: https://arxiv.org/abs/2407.12173

Add "beta" to the list of schedulers and the BetaSamplingScheduler node.
 7 months ago
喵哩个咪 855789403b
support clip-vit-large-patch14-336 (#4042) 
* support clip-vit-large-patch14-336

* support clip-vit-large-patch14-336
 7 months ago
comfyanonymous 6f7869f365 Get clip vision image size from config. 7 months ago
comfyanonymous 281ad42df4 Fix lowvram union controlnet bug. 7 months ago
Thomas Ward c5a48b15bd
Make default hash lib configurable without code changes via CLI argument (#3947) 
* cli_args: Add --duplicate-check-hash-function.

* server.py: compare_image_hash configurable hash function

Uses an argument added in cli_args to specify the type of hashing to default to for duplicate hash checking.  Uses an `eval()` to identify the specific hashlib class to utilize, but ultimately safely operates because we have specific options and only those options/choices in the arg parser.  So we don't have any unsafe input there.

* Add hasher() to node_helpers

* hashlib selection moved to node_helpers

* default-hashing-function instead of dupe checking hasher

This makes a default-hashing-function option instead of previous selected option.

* Use args.default_hashing_function

* Use safer handling for node_helpers.hasher()

Uses a safer handling method than `eval` to evaluate default hashing function.

* Stray parentheses are evil.

* Indentation fix.

Somehow when I hit save I didn't notice I missed a space to make indentation work proper.  Oops!
 7 months ago
comfyanonymous 8270c62530 Add SetUnionControlNetType to set the type of the union controlnet model. 7 months ago
comfyanonymous 821f93872e Allow model sampling to set number of timesteps. 7 months ago
Chenlei Hu 99458e8aca
Add `FrontendManager` to manage non-default front-end impl (#3897) 
* Add frontend manager

* Add tests

* nit

* Add unit test to github CI

* Fix path

* nit

* ignore

* Add logging

* Install test deps

* Remove 'stable' keyword support

* Update test

* Add web-root arg

* Rename web-root to front-end-root

* Add test on non-exist version number

* Use repo owner/name to replace hard coded provider list

* Inline cmd args

* nit

* Fix unit test
 7 months ago
comfyanonymous 1305fb294c Refactor: Move some code to the comfy/text_encoders folder. 7 months ago
comfyanonymous 7914c47d5a Quick fix for the promax controlnet. 7 months ago
comfyanonymous a3dffc447a Support AuraFlow Lora and loading model weights in diffusers format. 
You can load model weights in diffusers format using the UNETLoader node.
 8 months ago
comfyanonymous 29c2e26724 Better tokenizing code for AuraFlow. 8 months ago
comfyanonymous 8e012043a9 Add a ModelSamplingAuraFlow node to change the shift value. 
Set the default AuraFlow shift value to 1.73 (sqrt(3)).
 8 months ago
comfyanonymous 9f291d75b3 AuraFlow model implementation. 8 months ago
comfyanonymous f45157e3ac Fix error message never being shown. 8 months ago
comfyanonymous 5e1fced639 Cleaner support for loading different diffusion model types. 8 months ago
comfyanonymous ffe0bb0a33 Remove useless code. 8 months ago
comfyanonymous 391c1046cf More flexibility with text encoder return values. 
Text encoders can now return other values to the CONDITIONING than the cond
and pooled output.
 8 months ago
comfyanonymous e44fa5667f Support returning text encoder attention masks. 8 months ago
Extraltodeus f1a01c2c7e
Add sampler_pre_cfg_function (#3979) 
* Update samplers.py

* Update model_patcher.py
 8 months ago
comfyanonymous ade7aa1b0c Remove useless import. 8 months ago
comfyanonymous faa57430b0 Controlnet union model basic implementation. 
This is only the model code itself, it currently defaults to an empty
embedding [0] * 6 which seems to work better than treating it like a
regular controlnet.

TODO: Add nodes to select the image type.
 8 months ago
comfyanonymous bb663bcd6c Rename clip_t5base to t5base for stable audio text encoder. 8 months ago
comfyanonymous 2dc84d1444 Add a way to set the timestep multiplier in the flow sampling. 8 months ago
comfyanonymous ff63893d10 Support other types of T5 models. 8 months ago
comfyanonymous 4040491149 Better T5xxl detection. 8 months ago
comfyanonymous b8e58a9394 Cleanup T5 code a bit. 8 months ago
comfyanonymous 80c4590998 Allow specifying the padding token for the tokenizer. 8 months ago
comfyanonymous ce649d61c0 Allow zeroing out of embeds with unused attention mask. 8 months ago
comfyanonymous 739b76630e Remove useless code. 8 months ago
comfyanonymous d7484ef30c Support loading checkpoints with the UNETLoader node. 8 months ago
comfyanonymous 537f35c7bc Don't update dict if contiguous. 8 months ago
Alex "mcmonkey" Goodwin 3f46362d22
fix non-contiguous tensor saving (from channels-last) (#3932) 8 months ago