llama.cpp
a07c32ea - llama : use F32 precision in GLM4 attention and no FA (#9130)

Commit

1 year ago

llama : use F32 precision in GLM4 attention and no FA (#9130)

References

#9130 - llama:use F32 precision in GLM4 attention and no FA

Author

piDack

piDack

Parents

Loading