vrr/koboldcpp
2
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2025-09-15 19:39:42 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

fix unintended sd model quantization (#1672)

Browse source 
The recent ggml update added another quant type, GGML_TYPE_MXFP4,
which got the same value as SD_TYPE_COUNT. That made the embedded
sd.cpp quantize to GGML_TYPE_MXFP4 by default.

Photomaker in particular ends up crashing due to
"Missing CPY op for types: f32 mxfp4".
This commit is contained in:
Wagner Bruna 2025-08-07 23:19:58 -03:00 • committed by GitHub
parent 8f15461bea
commit eed5577aaa
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
2 changed files with 4 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

2
otherarch/sdcpp/stable-diffusion.h
View file

@ -100,7 +100,7 @@ enum sd_type_t {
SD_TYPE_IQ4_NL_4_4 = 36,
// SD_TYPE_IQ4_NL_4_8 = 37,
// SD_TYPE_IQ4_NL_8_8 = 38,
SD_TYPE_COUNT = 39,
SD_TYPE_COUNT = 40,
};
SD_API const char* sd_type_name(enum sd_type_t type);