Skip to content

fix: prevent F.linear from saving dequantized weights in MatMul4Bit/MatMul8bitLt to save ~13GB VRAM and prevent OOM errors#1935

Open
butterwecksolutions wants to merge 1 commit intobitsandbytes-foundation:mainfrom
butterwecksolutions:main
Open

fix: prevent F.linear from saving dequantized weights in MatMul4Bit/MatMul8bitLt to save ~13GB VRAM and prevent OOM errors#1935
butterwecksolutions wants to merge 1 commit intobitsandbytes-foundation:mainfrom
butterwecksolutions:main

Commits

Commits on May 2, 2026