Disty0 75926e6fc4 Remove sdnq_use_quantized_matmul and use sdnq_quantize_matmul_mode instead

2026-07-15 18:45:30 +03:00

..

kernels

Remove sdnq_use_quantized_matmul and use sdnq_quantize_matmul_mode instead

2026-07-15 18:45:30 +03:00

layers

cleanup sdnq

2026-07-14 20:55:47 +03:00

packed_int

lint

2026-06-02 17:13:08 +02:00

__init__.py

Refactor SDNQ quantizer handling and add modules_to_not_use_matmul

2026-05-15 05:47:32 +03:00

common.py

cleanup

2026-07-15 17:18:19 +03:00

dequantizer.py

upcast fp8 weights instead of disable compile

2026-07-14 17:22:37 +03:00

file_loader.py

full codespell coverage

2026-06-04 12:36:10 +02:00

forward.py

SDNQ add fused Triton kernels

2026-07-13 18:01:01 +03:00

kernel_wrappers.py

Enable Triton MM on all GPUs

2026-07-14 22:21:53 +03:00

loader.py

SDNQ add fused Triton kernels

2026-07-13 18:01:01 +03:00

packed_float.py

SDNQ handle subnormals in FP formats and cleanup Triton MM

2026-07-11 21:52:51 +03:00

quant_utils.py

SDNQ add fused Triton kernels

2026-07-13 18:01:01 +03:00

quantizer.py

SDNQ add fused Triton kernels

2026-07-13 18:01:01 +03:00

utils.py

Inductor breaks down with math.log

2026-07-09 14:06:02 +03:00