From 52ea1813898f0a809e16600ae5980bae8f0887c2 Mon Sep 17 00:00:00 2001 From: Vladimir Mandic Date: Mon, 2 Dec 2024 12:26:27 -0500 Subject: [PATCH] update requirements and notes Signed-off-by: Vladimir Mandic --- CHANGELOG.md | 9 +++++++-- TODO.md | 3 ++- requirements.txt | 2 +- 3 files changed, 10 insertions(+), 4 deletions(-) diff --git a/CHANGELOG.md b/CHANGELOG.md index 8091176e8..748ac6f04 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -29,13 +29,13 @@ style-aligned applies selected attention layers uniformly to all images to achive consistency can be used with or without input image in which case first prompt is used to establish baseline *note:* all prompts are processes as a single batch, so vram is limiting factor -- **OpenVINO**: update to 2024.5.0 ### UI and workflow improvements - **LoRA** handler rewrite: - LoRA weights are no longer calculated on-the-fly during model execution, but are pre-calculated at the start this results in perceived overhead on generate startup, but results in overall faster execution as LoRA does not need to be processed on each step + thanks @AI-Casanova - *note*: LoRA weights backups are required so LoRA can be unapplied, but can take quite a lot of system memory if you know you will not need to unapply LoRA, you can disable backups in *settings -> networks -> lora fuse* in which case, you need to reload model to unapply LoRA @@ -51,6 +51,7 @@ - faster and more compatible *balanced* mode - balanced offload: units are now in percentage instead of bytes - balanced offload: add both high and low watermark + *note*: balanced offload is recommended method for offload when using any large models such as sd35 or flux - **UI**: - improved stats on generate completion - improved live preview display and performance @@ -60,7 +61,11 @@ - control: optionn to hide input column - control: add stats - browser -> server logging framework - - add addtional themes: `black-reimagined` + - add addtional themes: `black-reimagined`, thanks @Artheriax + +### Updates + +- **OpenVINO**: update to 2024.5.0 - **Sampler** improvements - Euler FlowMatch: add sigma methods (*karras/exponential/betas*) - DPM FlowMatch: update all and add sigma methods diff --git a/TODO.md b/TODO.md index 73008039d..90372e41f 100644 --- a/TODO.md +++ b/TODO.md @@ -9,7 +9,8 @@ Main ToDo list can be found at [GitHub projects](https://github.com/users/vladma - Flux IPAdapter: - Flux NF4: - SANA: -- LTX-Video: +- LTX-Video: +- TorchAO: ## Other diff --git a/requirements.txt b/requirements.txt index 12a9f85cb..3b1b14c7d 100644 --- a/requirements.txt +++ b/requirements.txt @@ -52,7 +52,7 @@ numba==0.59.1 protobuf==4.25.3 pytorch_lightning==1.9.4 tokenizers==0.20.3 -transformers==4.46.2 +transformers==4.46.3 urllib3==1.26.19 Pillow==10.4.0 timm==0.9.16