aboutsummaryrefslogtreecommitdiff
path: root/modules/sd_samplers_common.py
AgeCommit message (Collapse)Author
2024-01-27Add Pad conds v0 optionAUTOMATIC1111
2023-08-30fix an issue where VAE would remain in fp16 after an auto-switch to fp32AUTOMATIC1111
2023-08-30add an option to choose how to combine hires fix and refinerAUTOMATIC1111
2023-08-22actual solution to the uncommon hanging problem that is seemingly caused by ↵AUTOMATIC1111
multiple progress requests working on same tensor
2023-08-22attemped solution to the uncommon hanging problem that is seemingly caused ↵AUTOMATIC1111
by live previews working on the tensor as denoising
2023-08-21forbid Full live preview method for medvram and add a setting to undo the ↵AUTOMATIC1111
forbidding
2023-08-19Merge pull request #12630 from catboxanon/fix/nans-mk2AUTOMATIC1111
Attempt to resolve NaN issue with unstable VAEs in fp32 mk2
2023-08-17Add option for faster live interruptcatboxanon
2023-08-17Attempt to resolve NaN issue with unstable VAEs in fp32 mk2catboxanon
2023-08-16Remove wrong scaleKohaku-Blueleaf
2023-08-14correctly add Eta DDIM to infotext when it's 1.0 and do not add it when it's ↵AUTOMATIC1111
0.0.
2023-08-13Merge pull request #12514 from catboxanon/feat/batch-encodeAUTOMATIC1111
Encode batch items individually to significantly reduce VRAM
2023-08-13Add s_noise param to more samplerscatboxanon
2023-08-13Encode batches separatelycatboxanon
Significantly reduces VRAM. This makes encoding more inline with how decoding currently functions.
2023-08-13use dataclass for StableDiffusionProcessingAUTOMATIC1111
2023-08-13add refiner to StableDiffusionProcessing classAUTOMATIC1111
write out correct model name in infotext, rather than the refiner model
2023-08-12make it possible to use hires fix together with refinerAUTOMATIC1111
2023-08-12put refiner into main UI, into the new accordions sectionAUTOMATIC1111
add VAE from main model into infotext, not from refiner model option to make scripts UI without gr.Group fix inconsistencies with refiner when usings samplers that do more denoising than steps
2023-08-10Merge branch 'dev' into refinerAUTOMATIC1111
2023-08-09rework RNG to use generators instead of generating noises beforehandAUTOMATIC1111
2023-08-08add None refiner optionAUTOMATIC1111
2023-08-08merge errorsAUTOMATIC1111
2023-08-08Merge branch 'dev' into refinerAUTOMATIC1111
2023-08-08add CFG denoiser implementation for DDIM, PLMS and UniPC (this is the commit ↵AUTOMATIC1111
when you can run both old and new implementations to compare them)
2023-08-06add infotextAUTOMATIC1111
add proper support for recalculating conds in k-diffusion samplers remove support for compvis samplers
2023-08-06initial refiner supportAUTOMATIC1111
2023-08-06repair broken live previews if using VAE with halfAUTOMATIC1111
2023-08-04remove noneed scaleKohaku-Blueleaf
2023-08-04Revert "change all encode"Kohaku-Blueleaf
This reverts commit 094c416a801b16c7d8e1944e2e9fae2c9e98bf12.
2023-08-04change all encodeKohaku-Blueleaf
2023-08-04move the modified decode into smapler_commonKohaku-Blueleaf
2023-08-04Merge branch 'dev' into efficient-vae-methodsKohaku-Blueleaf
2023-08-04resolve some of circular import issues for kohakuAUTOMATIC1111
2023-08-04Fix code styleKohaku-Blueleaf
2023-08-04add TAESD for i2i and t2iKohaku-Blueleaf
2023-08-03rework torchsde._brownian.brownian_interval replacement to use ↵AUTOMATIC1111
device.randn_local and respect the NV setting.
2023-05-17simplify single_sample_to_imageAUTOMATIC
2023-05-17TAESD fixSakura-Luna
2023-05-17Merge branch 'dev' into taesd-aAUTOMATIC1111
2023-05-17return live preview defaults to how they wereAUTOMATIC
only download TAESD model when it's needed return calculations in single_sample_to_image to just if/elif/elif blocks keep taesd model in its own directory
2023-05-16suppress ENSD infotext for samplers that don't use itAUTOMATIC
2023-05-14Add live preview mode checkSakura-Luna
2023-05-14Add Tiny AE live previewSakura-Luna
2023-04-29rename CPU RNG to RNG source in settings, add infotext and parameters ↵AUTOMATIC
copypaste support to RNG source
2023-04-18Option to use CPU for random number generation.Deciare
Makes a given manual seed generate the same images across different platforms, independently of the GPU architecture in use. Fixes #9613.
2023-02-01Refactor Mac specific code to a separate filebrkirch
Move most Mac related code to a separate file, don't even load it unless web UI is run under macOS.
2023-01-30split samplers into one more files for k-diffusionAUTOMATIC
2023-01-30split compvis sampler and shared sampler stuff into their own filesAUTOMATIC
2023-01-30Split history sd_samplers.py to sd_samplers_common.pyAndrey
generated by cgit v1.2.1 (git 2.18.0) at 2024-09-20 05:29:11 +0000