1338 Commits

Author SHA1 Message Date
comfyanonymous
ff7449847e Upcasting rope to fp32 seems to make no difference in this model. 2025-02-05 04:32:47 -05:00
comfyanonymous
f210ee69df Use regular numbers for rope in lumina model. 2025-02-05 04:17:25 -05:00
comfyanonymous
eaf581f1ed Fix lumina 2 batches. 2025-02-04 21:48:11 -05:00
comfyanonymous
0467641f12 Lower the default shift of lumina to reduce artifacts. 2025-02-04 06:50:37 -05:00
comfyanonymous
24b0e00898 Fix on python 3.9 2025-02-04 04:20:56 -05:00
comfyanonymous
e03a640306 Support Lumina 2 model. 2025-02-04 04:16:30 -05:00
comfyanonymous
fc1a24c402 Use maximum negative value instead of -inf for masks in text encoders.
This is probably more correct.
2025-02-02 09:46:00 -05:00
Dr.Lt.Data
4a3500c3ef better guide message for sageattention (#6634) 2025-02-02 09:26:47 -05:00
KarryCharon
e12555bdbf add disable-compres-response-body cli args; add compress middleware; (#6672) 2025-02-02 09:24:55 -05:00
comfyanonymous
af91112845 Only use stable cascade lora format with cascade model. 2025-02-01 06:35:22 -05:00
comfyanonymous
dcacbab9b0 Allow batch of different sigmas when noise scaling. 2025-01-30 06:49:52 -05:00
filtered
acf1f98438 Allow changing folder_paths.base_path via command line argument. (#6600)
* Reimpl. CLI arg directly inside folder_paths.

* Update tests to use CLI arg mocking.

* Revert last-minute refactor.

* Fix test state polution.
2025-01-29 08:06:28 -05:00
comfyanonymous
9dbb2b6d89 More friendly error messages for corrupted safetensors files. 2025-01-28 09:41:09 -05:00
comfyanonymous
5da1230892 Lower minimum ratio of loaded weights on Nvidia. 2025-01-27 05:26:51 -05:00
comfyanonymous
00e27c6202 Remove redundant code. 2025-01-25 19:04:53 -05:00
comfyanonymous
6da2db0e10 Remove useless code. 2025-01-24 06:15:54 -05:00
comfyanonymous
9cc63be0a0 Remove useless code. 2025-01-23 05:56:23 -05:00
Chenlei Hu
25f4e415da Remove unused function lcm in conds.py (#6572) 2025-01-23 05:54:09 -05:00
comfyanonymous
e3f3830dde Remove support for python 3.8. 2025-01-22 17:04:30 -05:00
chaObserv
4d5a8e8525 Add gradient estimation sampler (#6554) 2025-01-22 05:29:40 -05:00
comfyanonymous
eb64454c08 Add FluxDisableGuidance node to disable using the guidance embed. 2025-01-20 14:50:24 -05:00
comfyanonymous
71281dcc56 Cleanup old TODO. 2025-01-20 03:44:13 -05:00
Sergii Dymchenko
441dfbf8d9 Use torch.special.expm1 (#6388)
* Use `torch.special.expm1`

This function provides greater precision than `exp(x) - 1` for small values of `x`.

Found with TorchFix https://github.com/pytorch-labs/torchfix/

* Use non-alias
2025-01-19 04:54:32 -05:00
catboxanon
ccec4da418 Remove comfy.samplers self-import (#6506) 2025-01-18 17:49:51 -05:00
comfyanonymous
39b2c8eeee Uni pc sampler now works with audio and video models. 2025-01-18 05:27:58 -05:00
comfyanonymous
b8fbb7c807 Add warning when using old pytorch versions. 2025-01-17 18:47:27 -05:00
comfyanonymous
d439446067 Fix some cosmos fp8 issues. 2025-01-16 17:45:37 -05:00
comfyanonymous
63af9e9d90 Fix cosmos VAE failing with videos longer than 121 frames. 2025-01-16 16:30:06 -05:00
comfyanonymous
5650bfe1a9 Code refactor. 2025-01-16 07:23:54 -05:00
comfyanonymous
9ea569d2b0 Tweak hunyuan memory usage factor. 2025-01-16 06:31:03 -05:00
comfyanonymous
868a4d1551 Clean up some debug lines. 2025-01-16 04:24:39 -05:00
comfyanonymous
ad3ed4154c More accurate memory estimation for cosmos and hunyuan video. 2025-01-16 03:48:40 -05:00
comfyanonymous
220f470bfa Slightly lower hunyuan video memory usage. 2025-01-16 00:23:01 -05:00
comfyanonymous
3423ffcf70 Lower cosmos diffusion model memory usage. 2025-01-15 23:46:42 -05:00
comfyanonymous
d20c6506b5 Lower cosmos VAE memory usage by a bit. 2025-01-15 22:57:52 -05:00
comfyanonymous
8bf2059401 Optimize first attention block in cosmos VAE. 2025-01-15 21:48:46 -05:00
comfyanonymous
fd4d654129 Remove unsafe embedding load for very old pytorch. 2025-01-15 04:32:23 -05:00
comfyanonymous
34cf225065 Force safe loading of files in torch format on pytorch 2.4+
If this breaks something for you make an issue.
2025-01-15 03:50:27 -05:00
Pam
21ac610254 Rewrite res_multistep sampler and implement res_multistep_cfg_pp sampler. (#6462) 2025-01-14 18:20:06 -05:00
comfyanonymous
f19e3ec763 Implement Cosmos Image/Video to World (Video) diffusion models.
Use CosmosImageToVideoLatent to set the input image/video.
2025-01-14 05:14:10 -05:00
comfyanonymous
21a77abd0b Remove useless code. 2025-01-13 03:52:37 -05:00
comfyanonymous
302635bf4d Add res_multistep sampler from the cosmos code.
This sampler should work with all models.
2025-01-12 03:10:07 -05:00
Jedrzej Kosinski
ec567e2b7d Hooks Part 2 - TransformerOptionsHook and AdditionalModelsHook (#6377)
* Add 'sigmas' to transformer_options so that downstream code can know about the full scope of current sampling run, fix Hook Keyframes' guarantee_steps=1 inconsistent behavior with sampling split across different Sampling nodes/sampling runs by referencing 'sigmas'

* Cleaned up hooks.py, refactored Hook.should_register and add_hook_patches to use target_dict instead of target so that more information can be provided about the current execution environment if needed

* Refactor WrapperHook into TransformerOptionsHook, as there is no need to separate out Wrappers/Callbacks/Patches into different hook types (all affect transformer_options)

* Refactored HookGroup to also store a dictionary of hooks separated by hook_type, modified necessary code to no longer need to manually separate out hooks by hook_type

* In inner_sample, change "sigmas" to "sampler_sigmas" in transformer_options to not conflict with the "sigmas" that will overwrite "sigmas" in _calc_cond_batch

* Refactored 'registered' to be HookGroup instead of a list of Hooks, made AddModelsHook operational and compliant with should_register result, moved TransformerOptionsHook handling out of ModelPatcher.register_all_hook_patches, support patches in TransformerOptionsHook properly by casting any patches/wrappers/hooks to proper device at sample time

* Made hook clone code sane, made clear ObjectPatchHook and SetInjectionsHook are not yet operational

* Fix performance of hooks when hooks are appended via Cond Pair Set Props nodes by properly caching between positive and negative conds, make hook_patches_backup behave as intended (in the case that something pre-registers WeightHooks on the ModelPatcher instead of registering it at sample time)

* Filter only registered hooks on self.conds in CFGGuider.sample

* Make hook_scope functional for TransformerOptionsHook

* removed 4 whitespace lines to satisfy Ruff,

* Add a get_injections function to ModelPatcher

* Made TransformerOptionsHook contribute to registered hooks properly, added some doc strings and removed a so-far unused variable

* Rename AddModelsHooks to AdditionalModelsHook, rename SetInjectionsHook to InjectionsHook (not yet implemented, but at least getting the naming figured out)

* Clean up a typehint
2025-01-11 12:20:23 -05:00
comfyanonymous
e51102c5a8 Fast latent preview for Cosmos. 2025-01-11 04:41:24 -05:00
comfyanonymous
6d9f172aeb WIP support for Nvidia Cosmos 7B and 14B text to world (video) models. 2025-01-10 09:14:16 -05:00
comfyanonymous
efae999351 Add argument to skip the output reshaping in the attention functions. 2025-01-10 06:27:37 -05:00
comfyanonymous
f57d3080b2 Cleaner handling of attention mask in ltxv model code. 2025-01-09 07:12:03 -05:00
comfyanonymous
095a9d32c7 Improve some logging messages. 2025-01-08 19:05:22 -05:00
comfyanonymous
6b210b9e9b Properly calculate inner dim for t5 model.
This is required to support some different types of t5 models.
2025-01-07 17:33:03 -05:00
comfyanonymous
c75fa1bbf4 Make a few more samplers deterministic. 2025-01-07 02:12:32 -05:00