Block swap and fp8 scaled, fp8 already quantized warning #2224
kohya-ss
merged
11 commits into
kohya-ss:feat-flux-chroma-fp8-scaled
from
rockerBOO:block-swap-fp8-scaled
Supplement the input parameters to correctly convert the flux model t…
e64dc05c
feat: add new parameters for sample image inference configuration
6a826d21
Merge pull request #2212 from kohya-ss/fix-hunyuan-image-sample-gener…
67d06213
docs: enhance text encoder CPU usage instructions for HunyuanImage-2.…
a0c26a0e
Merge pull request #2213 from kohya-ss/doc-hunyuan-image-training-tex…
f0c767e0
fix: disable_mmap_safetensors not defined in SDXL TI training
60bfa97b
Merge pull request #2216 from kohya-ss/fix-sdxl-textual-inversion-tra…
206adb64
Merge branch 'main' into dev
63711390
Merge branch 'dev' into sd3
5462a6bb
Merge pull request #2003 from laolongboy/sd3-dev
5e366acd
Set is_swapping_blocks before loading_device, add warning for ignorin…
f9710863
Reviewers
No reviews
Assignees
No one assigned
Labels
None yet
Milestone
No milestone
Login to write a write a comment.
Login via GitHub