4 people 06d020bb6e [Bugfix] Fix SM121 (DGX Spark) exclusion from Marlin/CUTLASS FP8 paths (#35568) ...

Signed-off-by: Blake Ledden <blake@secondnaturecomputing.com>
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Co-authored-by: Pavani Majety <pmajety@nvidia.com>

2026-05-15 10:59:00 -07:00

..

marlin_moe_wna16

[Bugfix] Fix SM121 (DGX Spark) exclusion from Marlin/CUTLASS FP8 paths (#35568)

2026-05-15 10:59:00 -07:00

mxfp8_moe

[Kernel] Integrate SM100 MXFP8 blockscaled grouped MM and quant kernels (#34448)

2026-03-01 23:31:19 -08:00

permute_unpermute_kernels

[Compile] Fix compile warning in moe_permute (#36529)

2026-03-16 10:16:14 -04:00

dsv3_router_gemm_bf16_out.cu

[ModelBash][DSV3] Add TRTLLM DSV3 Router GEMM kernel (6% B1 Speedup) (#34302)

2026-02-23 14:02:26 +00:00

dsv3_router_gemm_entry.cu

[Bugfix] Fix DSV3 kernels breaking _C and _moe_C on unsupported arches (#35123)

2026-02-23 17:11:27 -08:00

dsv3_router_gemm_float_out.cu

[ModelBash][DSV3] Add TRTLLM DSV3 Router GEMM kernel (6% B1 Speedup) (#34302)

2026-02-23 14:02:26 +00:00

dsv3_router_gemm_utils.h

[ModelBash][DSV3] Add TRTLLM DSV3 Router GEMM kernel (6% B1 Speedup) (#34302)

2026-02-23 14:02:26 +00:00

dsv4_norm_router_gemm_entry.cu

[DSV4] Fuse norm and router for low latency scenario (#41263)

2026-05-14 05:11:02 -07:00

dsv4_norm_router_gemm_kernel.cu

[DSV4] Fuse norm and router for low latency scenario (#41263)

2026-05-14 05:11:02 -07:00

dsv4_norm_router_gemm.h

[DSV4] Fuse norm and router for low latency scenario (#41263)

2026-05-14 05:11:02 -07:00

dynamic_4bit_int_moe_cpu.cpp

[CPU]Parallelize over tokens in int4 moe (#29600)

2025-12-02 06:21:39 +00:00

grouped_topk_kernels.cu

[Kernel] Optimize grouped topk kernel (#34206)

2026-02-20 01:34:45 -08:00

moe_align_sum_kernels.cu

[Bugfix][CI] fix typos (#34934)

2026-03-05 17:05:46 +00:00

moe_ops.h

[DSV4] Fuse norm and router for low latency scenario (#41263)

2026-05-14 05:11:02 -07:00

moe_permute_unpermute_op.cu

[Compile] Fix compile warning in moe_permute (#36529)

2026-03-16 10:16:14 -04:00

moe_wna16_utils.h

pre-commit autoupdate (#17380)

2025-04-29 06:46:55 -07:00

moe_wna16.cu

[BugFix] Accuracy fix for llama4 int4 - improperly casted scales (#16801)

2025-04-17 22:13:29 -07:00

moeTopKFuncs.cuh

[Kernel] Optimize grouped topk kernel (#34206)

2026-02-20 01:34:45 -08:00

topk_softmax_kernels.cu

fix: clamp NaN/Inf in topk_softmax to prevent duplicate expert IDs (#39391)

2026-04-21 15:04:41 +04:00

topk_softplus_sqrt_kernels.cu

[Compile] Fix compile warning with topk softplus sqrt (#41261)

2026-05-14 05:12:50 -07:00

torch_bindings.cpp

[DSV4] Fuse norm and router for low latency scenario (#41263)

2026-05-14 05:11:02 -07:00