Files

 main

/

src

/

Copy path

Directory actions

More options

Directory actions

More options

Latest commit

happierpig

feat: Naive Support for Hopper FP8 Prefill Kernel with Per-Head Quant…

Feb 27, 2025

f5dec3d · Feb 27, 2025

History

History

This branch is 6 commits behind flashinfer-ai/flashinfer:main.

/

src

/

Top

Folders and files

Name		Name	Last commit message	Last commit date
parent directory ..
fp8-dev		fp8-dev	feat: Naive Support for Hopper FP8 Prefill Kernel with Per-Head Quant…	Feb 27, 2025
bench_batch_decode.cu		bench_batch_decode.cu	[Refactor] Unify JIT/Customization/AOT mode (flashinfer-ai#748)	Jan 23, 2025
bench_batch_decode_mla.cu		bench_batch_decode_mla.cu	[Refactor] Unify JIT/Customization/AOT mode (flashinfer-ai#748)	Jan 23, 2025
bench_batch_prefill.cu		bench_batch_prefill.cu	[Refactor] Unify JIT/Customization/AOT mode (flashinfer-ai#748)	Jan 23, 2025
bench_cascade.cu		bench_cascade.cu	[Refactor] Unify JIT/Customization/AOT mode (flashinfer-ai#748)	Jan 23, 2025
bench_norm.cu		bench_norm.cu	bench: add sampling & norm benchmarks (flashinfer-ai#208)	Apr 23, 2024
bench_sampling.cu		bench_sampling.cu	feat: more sampling operator options (flashinfer-ai#431)	Aug 9, 2024
bench_single_decode.cu		bench_single_decode.cu	[Refactor] Unify JIT/Customization/AOT mode (flashinfer-ai#748)	Jan 23, 2025
bench_single_prefill.cu		bench_single_prefill.cu	[Refactor] Unify JIT/Customization/AOT mode (flashinfer-ai#748)	Jan 23, 2025
cpu_reference.h		cpu_reference.h	perf: accelerate JIT compilation speed (flashinfer-ai#618)	Nov 20, 2024
flashinfer_ops.cuh		flashinfer_ops.cuh	feat: support deepseek prefill attention shape (flashinfer-ai#765)	Feb 1, 2025
test_attn_all_reduce.cu		test_attn_all_reduce.cu	feat: JIT compilation (flashinfer-ai#507)	Oct 7, 2024
test_batch_decode.cu		test_batch_decode.cu	[Refactor] Unify JIT/Customization/AOT mode (flashinfer-ai#748)	Jan 23, 2025
test_batch_prefill.cu		test_batch_prefill.cu	[Refactor] Unify JIT/Customization/AOT mode (flashinfer-ai#748)	Jan 23, 2025
test_cascade.cu		test_cascade.cu	[Refactor] Unify JIT/Customization/AOT mode (flashinfer-ai#748)	Jan 23, 2025
test_fast_dequant.cu		test_fast_dequant.cu	perf: faster fp8->fp16 dequantization for pre sm_90 arch (flashinfer-…	Aug 11, 2024
test_fastdiv.cu		test_fastdiv.cu	misc: suppress compilation warning of fastdiv (flashinfer-ai#279)	Jun 2, 2024
test_norm.cu		test_norm.cu	sampling: fix alignment issue for vocab_size not divisible by vec_size (	Apr 24, 2024
test_page.cu		test_page.cu	misc: Finer-grained control over fp16/fp8 builds (flashinfer-ai#722)	Jan 8, 2025
test_sampling.cu		test_sampling.cu	ci: setup pre-commit (flashinfer-ai#584)	Nov 5, 2024
test_single_decode.cu		test_single_decode.cu	misc: Finer-grained control over fp16/fp8 builds (flashinfer-ai#722)	Jan 8, 2025
test_single_prefill.cu		test_single_prefill.cu	[Refactor] Unify JIT/Customization/AOT mode (flashinfer-ai#748)	Jan 23, 2025
test_sum_all_reduce.cu		test_sum_all_reduce.cu	feat: initial support of distributed operators (flashinfer-ai#289)	Jun 8, 2024
utils.h		utils.h	[Refactor] Unify JIT/Customization/AOT mode (flashinfer-ai#748)	Jan 23, 2025
View all files