aboutsummaryrefslogtreecommitdiffstats
path: root/modules/sd_hijack_optimizations.py
AgeCommit message (Expand)AuthorLines
2023-08-13Make sub-quadratic the default for MPSbrkirch-2/+5
2023-08-13Use fixed size for sub-quadratic chunking on MPSbrkirch-1/+5
2023-08-02update doggettx cross attention optimization to not use an unreasonable amoun...AUTOMATIC1111-2/+2
2023-07-13get attention optimizations to workAUTOMATIC1111-7/+7
2023-07-12SDXL supportAUTOMATIC1111-8/+43
2023-06-07Merge pull request #11066 from aljungberg/patch-1AUTOMATIC1111-1/+1
2023-06-06Fix upcast attention dtype error.Alexander Ljungberg-1/+1
2023-06-04Merge pull request #10990 from vkage/sd_hijack_optimizations_bugfixAUTOMATIC1111-1/+1
2023-06-04fix the broken line for #10990AUTOMATIC-1/+1
2023-06-03torch.cuda.is_available() check for SdOptimizationXformersVivek K. Vasishtha-1/+1
2023-06-01revert default cross attention optimization to DoggettxAUTOMATIC-3/+3
2023-06-01revert default cross attention optimization to DoggettxAUTOMATIC-3/+3
2023-05-31rename print_error to report, use it with together with package nameAUTOMATIC-2/+1
2023-05-29Add & use modules.errors.print_error where currently printing exception info ...Aarni Koskela-4/+2
2023-05-21Add a couple `from __future__ import annotations`es for Py3.9 compatAarni Koskela-0/+1
2023-05-19Apply suggestions from code reviewAUTOMATIC1111-38/+28
2023-05-19fix linter issuesAUTOMATIC-1/+1
2023-05-18make it possible for scripts to add cross attention optimizationsAUTOMATIC-3/+132
2023-05-11Autofix Ruff W (not W605) (mostly whitespace)Aarni Koskela-16/+16
2023-05-10ruff auto fixesAUTOMATIC-7/+7
2023-05-10autofixes from ruffAUTOMATIC-1/+0
2023-05-08Fix for Unet NaNsbrkirch-0/+3
2023-03-24Update sd_hijack_optimizations.pyFNSpd-1/+1
2023-03-21Update sd_hijack_optimizations.pyFNSpd-1/+1
2023-03-10sdp_attnblock_forward hijackPam-0/+24
2023-03-10argument to disable memory efficient for sdpPam-0/+4
2023-03-07scaled dot product attentionPam-0/+42
2023-01-25Add UI setting for upcasting attention to float32brkirch-60/+99
2023-01-23better support for xformers flash attention on older versions of torchAUTOMATIC-24/+18
2023-01-21add --xformers-flash-attention option & implTakuma Mori-2/+24
2023-01-21extra networks UIAUTOMATIC-5/+5
2023-01-06Added licensebrkirch-0/+1
2023-01-06Change sub-quad chunk threshold to use percentagebrkirch-9/+9
2023-01-06Add Birch-san's sub-quadratic attention implementationbrkirch-25/+99
2022-12-20Use other MPS optimization for large q.shape[0] * q.shape[1]brkirch-4/+6
2022-12-10cleanup some unneeded imports for hijack filesAUTOMATIC-3/+0
2022-12-10do not replace entire unet for the resolution hackAUTOMATIC-28/+0
2022-11-23Patch UNet Forward to support resolutions that are not multiples of 64Billy Cao-0/+31
2022-10-19Remove wrong self reference in CUDA support for invokeaiCheka-1/+1
2022-10-18Update sd_hijack_optimizations.pyC43H66N12O12S2-0/+3
2022-10-18readd xformers attnblockC43H66N12O12S2-0/+15
2022-10-18delete xformers attnblockC43H66N12O12S2-12/+0
2022-10-11Use apply_hypernetwork functionbrkirch-10/+4
2022-10-11Add InvokeAI and lstein to credits, add back CUDA supportbrkirch-0/+13
2022-10-11Add check for psutilbrkirch-4/+15
2022-10-11Add cross-attention optimization from InvokeAIbrkirch-0/+79
2022-10-11rename hypernetwork dir to hypernetworks to prevent clash with an old filenam...AUTOMATIC-1/+1
2022-10-11fixes related to mergeAUTOMATIC-1/+2
2022-10-11replace duplicate code with a functionAUTOMATIC-29/+15
2022-10-10remove functorchC43H66N12O12S2-2/+0
generated by cgit v1.2.3 (git 2.39.1) at 2025-10-07 16:44:36 +0000