Skip to content

Conversation

wbruna
Copy link
Contributor

@wbruna wbruna commented Aug 7, 2025

Recently, GGML_TYPE_COUNT got bumped for the new GGML_TYPE_MXFP4 quant, getting it out-of-sync with SD_TYPE_COUNT.

To make it easier to build stable-diffusion.cpp against different ggml versions, adjust the type conversions to consider both GGML_TYPE_COUNT and SD_TYPE_COUNT as limits.

(this got committed on top of #716 by mistake; feel free to drop the first commit)

Recently, GGML_TYPE_COUNT got bumped for the new GGML_TYPE_MXFP4
quant, getting it out-of-sync with SD_TYPE_COUNT.

To make it easier to build stable-diffusion.cpp against different
ggml versions when other weight types get added to ggml, adjust
the type conversions to consider both GGML_TYPE_COUNT and
SD_TYPE_COUNT as limits.
@wbruna wbruna force-pushed the sd_weight_max_value branch from f9a6c7f to 1480d81 Compare September 13, 2025 19:03
@leejet leejet merged commit 5869987 into leejet:master Sep 14, 2025
8 checks passed
@leejet
Copy link
Owner

leejet commented Sep 14, 2025

Thank you for your contribution.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants