* Conv2DDirect for VAE stage
* Enable only for Vulkan, reduced duplicated code
* Cmake option to use conv2d direct
* conv2d direct always on for opencl
* conv direct as a flag
* fix merge typo
* Align conv2d behavior to flash attention's
* fix readme
* add conv2d direct for controlnet
* add conv2d direct for esrgan
* clean code, use enable_conv2d_direct/get_all_blocks
* format code
---------
Co-authored-by: leejet <leejet714@gmail.com>
* feat: add new scheduler types, clip skip and vae to image embedded params
- If a non default scheduler is set, include it in the 'Sampler' tag in the data
embedded into the final image.
- If a custom VAE path is set, include the vae name (without path and extension)
in embedded image params under a `VAE:` tag.
- If a custom Clip skip is set, include that Clip skip value in embedded image
params under a `Clip skip:` tag.
* feat: add separate diffusion and text models to metadata
---------
Co-authored-by: one-lithe-rune <skapusniak@lithe-runes.com>
* Instruct-p2p support
* support 2 conditionings cfg
* Do not re-encode the exact same image twice
* fixes for 2-cfg
* Fix pix2pix latent inputs + improve inpainting a bit + fix naming
* prepare for other pix2pix-like models
* Support sdxl ip2p
* fix reference image embeddings
* Support 2-cond cfg properly in cli
* fix typo in help
* Support masks for ip2p models
* unify code style
* delete unused code
* use edit mode
* add img_cond
* format code
---------
Co-authored-by: leejet <leejet714@gmail.com>
* repair flash attention in _ext
this does not fix the currently broken fa behind the define, which is only used by VAE
Co-authored-by: FSSRepo <FSSRepo@users.noreply.github.com>
* make flash attention in the diffusion model a runtime flag
no support for sd3 or video
* remove old flash attention option and switch vae over to attn_ext
* update docs
* format code
---------
Co-authored-by: FSSRepo <FSSRepo@users.noreply.github.com>
Co-authored-by: leejet <leejet714@gmail.com>
* mmdit-x
* add support for sd3.5 medium
* add skip layer guidance support (mmdit only)
* ignore slg if slg_scale is zero (optimization)
* init out_skip once
* slg support for flux (expermiental)
* warn if version doesn't support slg
* refactor slg cli args
* set default slg_scale to 0 (oops)
* format code
---------
Co-authored-by: leejet <leejet714@gmail.com>
* add flux support
* avoid build failures in non-CUDA environments
* fix schnell support
* add k quants support
* add support for applying lora to quantized tensors
* add inplace conversion support for f8_e4m3 (#359)
in the same way it is done for bf16
like how bf16 converts losslessly to fp32,
f8_e4m3 converts losslessly to fp16
* add xlabs flux comfy converted lora support
* update docs
---------
Co-authored-by: Erik Scholz <Green-Sky@users.noreply.github.com>
