feat: add flux support (#356)

Browse Source

* 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>

...

This commit is contained in:

leejet

2024-08-24 14:29:52 +08:00

committed by GitHub

parent 697d000f49

commit 64d231f384

25 changed files with 1886 additions and 172 deletions

Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL

Download Patch File Download Diff File Expand all files Collapse all files

BIN

assets/flux/flux1-dev-q2_k.png Normal file

Binary file not shown.

Side by Side After Width:  |  Height:  |  Size: 416 KiB

BIN

assets/flux/flux1-dev-q3_k.png Normal file

Binary file not shown.

Side by Side After Width:  |  Height:  |  Size: 490 KiB

BIN

assets/flux/flux1-dev-q4_0.png Normal file

Binary file not shown.

Side by Side After Width:  |  Height:  |  Size: 464 KiB

BIN

assets/flux/flux1-dev-q8_0 with lora.png Normal file

Binary file not shown.

Side by Side After Width:  |  Height:  |  Size: 566 KiB

BIN

assets/flux/flux1-dev-q8_0.png Normal file

Binary file not shown.

Side by Side After Width:  |  Height:  |  Size: 475 KiB

BIN

assets/flux/flux1-schnell-q8_0.png Normal file

Binary file not shown.

Side by Side After Width:  |  Height:  |  Size: 481 KiB