Skip to content

Commit 6897c60

Browse files
Isotr0pyIsotr0py
committed
ooops
Signed-off-by: Isotr0py <[email protected]>
1 parent 665aacd commit 6897c60

File tree

2 files changed

+1
-31
lines changed

2 files changed

+1
-31
lines changed

‎run_gguf.py‎

Lines changed: 0 additions & 30 deletions
This file was deleted.

‎src/diffusers/quantizers/gguf/utils.py‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -74,7 +74,7 @@
7474
# Consolidate DEQUANT_TYPES and MMQ_QUANT_TYPES after we add
7575
# MMQ kernel for I-Matrix quantization.
7676
DEQUANT_TYPES = STANDARD_QUANT_TYPES | KQUANT_TYPES | IMATRIX_QUANT_TYPES
77-
MMQ_QUANT_TYPES = STANDARD_QUANT_TYPES | KQUANT_TYPES
77+
MMQ_QUANT_TYPES = STANDARD_QUANT_TYPES
7878

7979

8080
def _fused_mul_mat_gguf(x: torch.Tensor, qweight: torch.Tensor, qweight_type: int) -> torch.Tensor:

0 commit comments

Comments
 (0)