leejet f5df5f8b84 perf: accelerate packed INT4 CUDA inference with CUTLASS (#2278)

2026-11-06 01:59:05 +08:00

..

sageattention

fix: include missing cassert headers in SageAttention CUDA code

2026-11-02 00:17:49 +08:00

affine.cu

refactor: move model-specific operators and fusions into backend extensions (#2253)

2026-11-01 21:18:37 +08:00

bf16_gemm.cu

perf: fuse Z-Image GPU operations and optimize BF16 GEMM (#2265)

2026-11-03 23:59:21 +08:00

bf16_gemm.h

perf: fuse Z-Image GPU operations and optimize BF16 GEMM (#2265)

2026-11-03 23:59:21 +08:00

bf16.cu

perf: fuse Z-Image GPU operations and optimize BF16 GEMM (#2265)

2026-11-03 23:59:21 +08:00

bf16.h

perf: fuse Z-Image GPU operations and optimize BF16 GEMM (#2265)

2026-11-03 23:59:21 +08:00

CMakeLists.txt

perf: accelerate packed INT4 CUDA inference with CUTLASS (#2278)

2026-11-06 01:59:05 +08:00

convrot_kernel.h

feat: add custom modulate operator and ConvRot fusion (#2258)

2026-11-02 01:19:28 +08:00

convrot.cu

feat: add custom modulate operator and ConvRot fusion (#2258)

2026-11-02 01:19:28 +08:00

cute_attention.cu

refactor: move model-specific operators and fusions into backend extensions (#2253)

2026-11-01 21:18:37 +08:00

i8_gated_residual.cu

refactor: move model-specific operators and fusions into backend extensions (#2253)

2026-11-01 21:18:37 +08:00

i8_gemm.cu

perf: accelerate packed INT4 CUDA inference with CUTLASS (#2278)

2026-11-06 01:59:05 +08:00

i8_gemm.h

perf: accelerate packed INT4 CUDA inference with CUTLASS (#2278)

2026-11-06 01:59:05 +08:00

i8_projection.cu

refactor: move model-specific operators and fusions into backend extensions (#2253)

2026-11-01 21:18:37 +08:00

i8_projection.h

refactor: move model-specific operators and fusions into backend extensions (#2253)

2026-11-01 21:18:37 +08:00

int8.cu

feat: support packed INT4 convrot models (#2276)

2026-11-06 00:35:58 +08:00

modulate.cu

feat: add custom modulate operator and ConvRot fusion (#2258)

2026-11-02 01:19:28 +08:00

registration.cu

feat: support packed INT4 convrot models (#2276)

2026-11-06 00:35:58 +08:00

registration.h

refactor: move model-specific operators and fusions into backend extensions (#2253)

2026-11-01 21:18:37 +08:00

rms_affine.cu

perf(cuda): fuse MOSS modulation and direct RoPE F16 output (#2269)

2026-11-04 02:01:41 +08:00

rms_affine.h

perf: fuse Z-Image GPU operations and optimize BF16 GEMM (#2265)

2026-11-03 23:59:21 +08:00

rope.cu

perf(cuda): fuse MOSS modulation and direct RoPE F16 output (#2269)

2026-11-04 02:01:41 +08:00

rope.h

refactor: move model-specific operators and fusions into backend extensions (#2253)

2026-11-01 21:18:37 +08:00

runtime.h

refactor: move model-specific operators and fusions into backend extensions (#2253)

2026-11-01 21:18:37 +08:00

sage_attention.cu

refactor: move model-specific operators and fusions into backend extensions (#2253)

2026-11-01 21:18:37 +08:00

sol_attention.cu

refactor: move model-specific operators and fusions into backend extensions (#2253)

2026-11-01 21:18:37 +08:00

w4_mma.h

perf: accelerate packed INT4 CUDA inference with CUTLASS (#2278)

2026-11-06 01:59:05 +08:00

w4.cu

perf: accelerate packed INT4 CUDA inference with CUTLASS (#2278)

2026-11-06 01:59:05 +08:00