PR #14810 convert : handle pre-quantized models

convert : begin handling pre-quantized models

compilade committed 248 days ago

Merge branch 'master' into compilade/convert-prequant

compilade committed 225 days ago

convert : fix conversion from FP8 for Deepseek-V3.1-Base

compilade committed 220 days ago

Merge branch 'master' into compilade/convert-prequant

compilade committed 207 days ago

Merge branch 'master' into compilade/convert-prequant

compilade committed 199 days ago

Merge branch 'master' into compilade/convert-prequant

compilade committed 155 days ago

llama.cpp convert : handle pre-quantized models #14810 Merged