aboutsummaryrefslogtreecommitdiffstats
path: root/modules/sd_samplers.py
Commit message (Collapse)AuthorAgeFilesLines
* Rework MPS randn fix, add randn_like fixbrkirch2022-11-301-3/+5
| | | | torch.manual_seed() already sets a CPU generator, so there is no reason to create a CPU generator manually. torch.randn_like also needs a MPS fix for k-diffusion, but a torch hijack with randn_like already exists so it can also be used for that.
* Add workaround for using MPS with torchsdebrkirch2022-11-301-0/+14
|
* rework #5012 to also work for pictures dragged into the prompt and also add ↵AUTOMATIC2022-11-271-1/+1
| | | | Clip skip + ENSD to parameters
* Merge pull request #5065 from JaySmithWpg/vram-leakAUTOMATIC11112022-11-271-18/+15
|\ | | | | #3449 - VRAM leak when switching to/from inpainting checkpoint
| * fixes #3449 - VRAM leak when switching to/from inpainting modelJay Smith2022-11-261-18/+15
| |
* | calculate dictionary for sampler names only onceAUTOMATIC2022-11-271-0/+7
| |
* | Merge branch 'master' into DPM++SDEAUTOMATIC11112022-11-271-6/+8
|\ \
| * | Add support Stable Diffusion 2.0AUTOMATIC2022-11-261-6/+8
| |/
* / Add DPM++ SDE sampleruservar2022-11-221-0/+2
|/
* change StableDiffusionProcessing to internally use sampler name instead of ↵AUTOMATIC2022-11-191-3/+10
| | | | sampler index
* tweak names a bit for new samplersAUTOMATIC2022-11-051-4/+4
|
* Reorder samplershentailord85ez2022-11-041-2/+2
|
* Fix namehentailord85ez2022-11-041-1/+1
|
* Add support for new DPM-Solver++ samplershentailord85ez2022-11-041-0/+4
|
* do not unnecessarily run VAE one more time when saving intermediate image ↵AUTOMATIC2022-11-021-0/+1
| | | | with hires fix
* Allow saving "before-highres-fix. (#4150)timntorres2022-11-021-3/+2
| | | * Save image/s before doing highres fix.
* Merge pull request #4021 from AUTOMATIC1111/add-kdiff-cfgdenoiser-callbackAUTOMATIC11112022-11-021-0/+7
|\ | | | | Add mid-kdiffusion cfgdenoiser script callback - access latents, conditionings and sigmas mid-sampling
| * Update class name and assign back to varsDepFA2022-11-021-2/+6
| |
| * Add callback to sd_samplersDepFA2022-10-301-0/+3
| |
* | Include PLMS in adjust steps as it also can fail in the same wayMartin Cairns2022-10-301-1/+1
| |
* | Add adjust_steps_if_invalid to find next valid step for ddim uniform samplerMartin Cairns2022-10-291-13/+15
|/
* fix #3145 #3093ritosonn2022-10-251-2/+2
|
* removed the option to use 2x more memory when generating previewsAUTOMATIC2022-10-221-25/+10
| | | | | added an option to always only show one image in previews removed duplicate code
* Generate grid preview for progress imageUnnoen2022-10-221-1/+25
|
* Merge branch 'master' into masterAUTOMATIC11112022-10-211-11/+49
|\
| * Made dummy latents smaller. Minor code cleanupsrandom_thoughtss2022-10-201-2/+4
| |
| * Don't compute latent mask if were not using it. Also added support for fixed ↵random_thoughtss2022-10-191-0/+4
| | | | | | | | highres_fix generation.
| * added support for ddim img2imgrandom_thoughtss2022-10-191-0/+6
| |
| * Added support for RunwayML inpainting modelrandom_thoughtss2022-10-191-12/+38
| |
* | Different approach to skip/interrupt with highres fixVladimir Repin2022-10-201-0/+4
|/
* remove dependence on TQDM for sampler progress/interrupt functionalityAUTOMATIC2022-10-181-49/+58
|
* Fix typo in commentsMartin Cairns2022-10-111-1/+1
|
* Remove debug code for checking that first sigma value is same after code cleanupMartin Cairns2022-10-111-1/+0
|
* Handle different parameters for DPM fast & adaptiveMartin Cairns2022-10-111-7/+18
|
* --no-half-vaeAUTOMATIC2022-10-101-2/+2
|
* chore: Fix typosAidan Holland2022-10-081-2/+2
|
* fix bugs related to variable prompt lengthsAUTOMATIC2022-10-081-7/+28
|
* check specifically for skippedTrung Ngo2022-10-081-2/+2
|
* karras samplers for img2img?AUTOMATIC2022-10-061-2/+4
|
* integrate the new samplers PRAUTOMATIC2022-10-061-28/+31
|
* add karras scheduling variantsC43H66N12O12S22022-10-061-0/+13
|
* option to let users select which samplers they want to hideAUTOMATIC2022-10-061-2/+17
|
* added support for AND from ↵AUTOMATIC2022-10-051-10/+25
| | | | https://energy-based-model.github.io/Compositional-Visual-Generation-with-Composable-Diffusion-Models/
* Fixed when eta = 0hentailord85ez2022-10-031-1/+1
| | | Unexpected behavior when using eta = 0 in something like XY, but your default eta was set to something not 0.
* added --disable-console-progressbars to disable progressbars in consoleAUTOMATIC2022-10-021-2/+6
| | | | disabled printing prompts to console by default, enabled by --enable-console-prompts
* add sampler_noise_scheduler_override switchDepFA2022-10-011-2/+8
|
* fix the sys not found error when importing sd_samplers that appeared after PRAUTOMATIC2022-09-301-2/+0
|
* force usage of cloned k-diffusionC43H66N12O12S22022-09-301-1/+2
|
* improve code qualityC43H66N12O12S22022-09-291-9/+7
|
* add new samplersC43H66N12O12S22022-09-291-1/+10
|