Cyrus LeungandGitHub 24d5186138 [Bugfix] Re-enable FP8 MoE on NVIDIA Thor (#46339) ...

Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>

2026-06-24 07:35:46 -07:00

..

c3x

[12/n] final _C library kernel migration (#45415)

2026-06-19 06:57:26 -07:00

moe

[10/n] Migrate cuda_view and silu_and_mul_per_block_quant kernels to torch stale ABI. (#44334)

2026-06-04 20:14:43 -07:00

scaled_mm_c2x_sm75_dispatch.cuh

[3/n] Migrate cutlass/scaled_mm_entry.cu torch stable ABI (#37221)

2026-03-30 11:20:13 -07:00

scaled_mm_c2x_sm80_dispatch.cuh

[3/n] Migrate cutlass/scaled_mm_entry.cu torch stable ABI (#37221)

2026-03-30 11:20:13 -07:00

scaled_mm_c2x_sm89_fp8_dispatch.cuh

[Perf] Batch invariance with Cutlass fp8 support, 28.9% E2E latency improvement (#40408)

2026-05-11 12:20:58 -04:00

scaled_mm_c2x_sm89_int8_dispatch.cuh

[3/n] Migrate cutlass/scaled_mm_entry.cu torch stable ABI (#37221)

2026-03-30 11:20:13 -07:00

scaled_mm_c2x.cu

[Perf] Batch invariance with Cutlass fp8 support, 28.9% E2E latency improvement (#40408)

2026-05-11 12:20:58 -04:00

scaled_mm_c2x.cuh

[12/n] final _C library kernel migration (#45415)

2026-06-19 06:57:26 -07:00

scaled_mm_c3x_sm90.cu

[3/n] Migrate cutlass/scaled_mm_entry.cu torch stable ABI (#37221)

2026-03-30 11:20:13 -07:00

scaled_mm_c3x_sm100.cu

[3/n] Migrate cutlass/scaled_mm_entry.cu torch stable ABI (#37221)

2026-03-30 11:20:13 -07:00

scaled_mm_c3x_sm120.cu

[3/n] Migrate cutlass/scaled_mm_entry.cu torch stable ABI (#37221)

2026-03-30 11:20:13 -07:00

scaled_mm_entry.cu

[Bugfix] Re-enable FP8 MoE on NVIDIA Thor (#46339)

2026-06-24 07:35:46 -07:00