Michael Goin d83f3f7cb3

Fixes and updates to bench_per_token_quant_fp8 (#25591) ...

Signed-off-by: Michael Goin <mgoin64@gmail.com>

2025-09-24 08:30:15 -07:00

..

deepgemm

[Performance] Move apply_w8a8_block_fp8_linear to an op class (#24666)

2025-09-23 12:03:10 -07:00

bench_block_fp8_gemm.py

[Perf] Use upstream CUTLASS for SM90 Block FP8 kernel (#23280)

2025-09-11 15:43:14 -07:00

bench_fp8_gemm.py

[Misc] Add SPDX-FileCopyrightText (#20428)

2025-07-04 07:40:42 +00:00

bench_int8_gemm.py

[Benchmark] Refactor benchmark script for fp8 & int8 (#19627)

2025-06-15 15:15:37 +08:00

bench_nvfp4_gemm.py

[Bench] Add NVFP4 GEMM benchmark script (#20578)

2025-07-09 13:23:48 -04:00

bench_per_token_quant_fp8.py

Fixes and updates to bench_per_token_quant_fp8 (#25591)

2025-09-24 08:30:15 -07:00

benchmark_activation.py

[Benchmark] add benchmark for custom activation op (#23908)

2025-09-06 20:12:05 -07:00

benchmark_bitblas.py

[Bugfix] Add proper comparison for package versions (#22314)

2025-08-06 20:31:03 -07:00

benchmark_cutlass_fp4_moe.py

[Kernel] Delegate construction of FusedMoEQuantConfig to FusedMoEMethodBase subclasses (#22537)

2025-09-17 17:43:31 -06:00

benchmark_cutlass_moe_fp8.py

Add CUTLASS FP8 MOE benchmark scripts and kernel config (#25302)

2025-09-23 18:07:42 -06:00

benchmark_device_communicators.py

[core] add nccl symmetric memory for all reduce (#24532)

2025-09-23 14:33:06 -04:00

benchmark_grouped_gemm_cutlass.py

[Kernel] Delegate construction of FusedMoEQuantConfig to FusedMoEMethodBase subclasses (#22537)

2025-09-17 17:43:31 -06:00

benchmark_layernorm.py

[Misc] Add SPDX-FileCopyrightText (#19100)

2025-06-03 11:20:17 -07:00

benchmark_lora.py

[fix] lora benchmarks pass no_lora_flag_cpu (#23774)

2025-09-17 21:22:25 +08:00

benchmark_machete.py

[kernel] Support W4A8 on Hopper (#23198)

2025-08-24 06:18:04 +00:00

benchmark_marlin.py

[Bugfix][Benchmark] Fix Marlin benchmark (#19929)

2025-06-24 07:25:12 +09:00

benchmark_moe_align_block_size.py

[Refactor] Remove moe_align_block_size_triton (#21335)

2025-07-26 07:08:29 -07:00

benchmark_moe_permute_unpermute.py

Fix CUDA permute/unpermute for use with DeepGemm Moe (#17934)

2025-07-27 07:08:00 -07:00

benchmark_moe.py

[Kernel] Delegate construction of FusedMoEQuantConfig to FusedMoEMethodBase subclasses (#22537)

2025-09-17 17:43:31 -06:00

benchmark_mrope.py

[FEAT] [Performance] Add triton mrope to replace the torch code path (#22375)

2025-08-09 11:50:03 -07:00

benchmark_paged_attention.py

[Misc] Add SPDX-FileCopyrightText (#19100)

2025-06-03 11:20:17 -07:00

benchmark_per_token_group_quant.py

[Test] Add Benchmark and Unit Test for per_token_group_quant (#21860)

2025-07-30 07:15:02 -07:00

benchmark_polynorm.py

[Model] New model support for Motif-1-Tiny (#23414)

2025-09-10 23:29:40 -07:00

benchmark_quant.py

[Misc] Add SPDX-FileCopyrightText (#19100)

2025-06-03 11:20:17 -07:00

benchmark_reshape_and_cache_flash.py

[V1][Kernel] Add triton implementation for reshape_and_cache_flash (#24503)

2025-09-23 12:52:40 -04:00

benchmark_rmsnorm.py

[Misc] Add SPDX-FileCopyrightText (#19100)

2025-06-03 11:20:17 -07:00

benchmark_rope.py

[Misc] Add SPDX-FileCopyrightText (#19100)

2025-06-03 11:20:17 -07:00

benchmark_shapes.py

[Misc] Add SPDX-FileCopyrightText (#19100)

2025-06-03 11:20:17 -07:00

benchmark_silu_mul_fp8_quant.py

[Kernels][DP/EP] Optimize Silu Kernel for R1 (#24054)

2025-09-13 00:17:27 -07:00

benchmark_trtllm_decode_attention.py

[Flashinfer] Support Flashinfer TRTLLM FP8-qkv BF16/FP16-out Attention Kernel (#23647)

2025-09-08 20:53:07 -07:00

benchmark_trtllm_prefill_attention.py

[Flashinfer] Support Flashinfer TRTLLM FP8-qkv BF16/FP16-out Attention Kernel (#23647)

2025-09-08 20:53:07 -07:00

benchmark_w8a8_block_fp8.py

[Frontend] Pass API server count to each process (#23717)

2025-09-20 01:15:19 +08:00

graph_machete_bench.py

[Misc] Add SPDX-FileCopyrightText (#19100)

2025-06-03 11:20:17 -07:00

requirements.txt

[Kernel] (2/N) Machete - Integrate into CompressedTensorsWNA16 and GPTQMarlin (#7701)

2024-09-23 13:46:26 -04:00

utils.py

[Misc] Add SPDX-FileCopyrightText (#19100)

2025-06-03 11:20:17 -07:00

weight_shapes.py

[kernel] Support W4A8 on Hopper (#23198)

2025-08-24 06:18:04 +00:00