7 people 9bd02dc5b9 feat(sgl-kernel): add InfLLM v2 attention kernels (#29383) ...

Co-authored-by: Size Wang <paulgeorge13hhhhh@gmail.com>
Co-authored-by: lijiayi <lijiayi@modelbest.cn>
Co-authored-by: suhmily10 <suhmily@gmail.com>
Co-authored-by: Xiaoyue Xu <xiaoyue.xu.me@gmail.com>
Co-authored-by: hansjohn <74091612+hansjohn@users.noreply.github.com>
Co-authored-by: zhangyan <1762895426@qq.com>

2026-07-06 22:46:54 -07:00

..

infllm_v2

feat(sgl-kernel): add InfLLM v2 attention kernels (#29383)

2026-07-06 22:46:54 -07:00

quantization

[7/n] decouple quantization impl from vllm dependency - gguf kernel (#11019)

2025-10-11 14:04:57 -07:00

testing

[kernel slimming] Clean many useless sgl-kernel deprecated kernels (#20277)

2026-03-14 16:45:54 +08:00

__init__.py

feat(sgl-kernel): add InfLLM v2 attention kernels (#29383)

2026-07-06 22:46:54 -07:00

allreduce.py

MSCCL++ Integration (#22734)

2026-06-08 21:13:13 -07:00

attention.py

Remove obsolete sgl-kernel legacy paths (#21528)

2026-04-01 09:00:20 +08:00

cutlass_moe.py

[1/n]: add cutlass W4A8 moe kernel for hopper architecture (#7772)

2025-07-04 20:50:12 -07:00

debug_utils.py

Add SGLang CUDA crash API logging inspired by FlashInfer (#20910)

2026-03-22 16:39:40 +08:00

elementwise.py

Revert "Fix wrong RMSNorm fallback to old Flashinfer CUDA kernel when in PCG" (#30083)

2026-07-03 19:05:20 -07:00

expert_specialization.py

[sgl-kernel][Feat][B200][1/N] Support MXFP8 Grouped GEMM in Blackwell (#13731)

2025-12-04 10:09:09 +08:00

flash_attn.py

Upgrade fa3 hash (#28394)

2026-06-17 13:32:11 -07:00

flash_mla.py

Sgl flashmla (#26132)

2026-05-26 12:00:23 -07:00

gemm.py

[JIT Kernel] Migrate dsv3_router_gemm from AOT sgl-kernel to JIT kernel (#21531)

2026-06-26 11:52:18 -07:00

grammar.py

fix sgl-kernel unit tests (#5666)

2025-04-23 01:18:30 -07:00

kvcacheio.py

[HiCache]Page head layout IO kernel (#11615)

2025-10-26 15:53:50 +08:00

load_utils.py

Update kernel installation instructions after shifting default cuda to 13 (#24181)

2026-05-02 16:07:20 -07:00

mamba.py

[CPU] add indices in chunk_gated_delta_rule (#29267)

2026-06-26 07:51:07 +08:00

memory.py

[kernel slimming] Clean many useless sgl-kernel deprecated kernels (#20277)

2026-03-14 16:45:54 +08:00

metal.py

chore: cleanup garbage code (#29770)

2026-07-02 16:14:01 +08:00

moe.py

[JIT Kernel] Reland NVFP4 kernels to JIT (#20012)

2026-03-07 10:31:08 +08:00

musa.py

[MUSA]: Add flashinfer sampling backend (#24978)

2026-05-14 20:23:15 -07:00

sampling.py

Remove obsolete sgl-kernel legacy paths (#21528)

2026-04-01 09:00:20 +08:00

scalar_type.py

refactor: Move scalar_types.py to sgl-kernel to avoid circular import (#8720)

2025-08-07 19:22:16 -07:00

sparse_flash_attn.py

[Feat] QWen-1M context support[1/2]: Update block sparse attention backend utils kernel (#5847)

2025-04-28 11:03:17 -07:00

spatial.py

Optional extension for green context (#9231)

2025-08-15 21:33:52 +08:00

speculative.py

move all get_stream in sgl_kernel to c++ to reduce the launch overhead (#12521)

2025-11-02 13:15:05 -08:00

test_utils.py

Reland [1/2] Optimizations and refactors about quant kernel (#10312)

2025-10-11 15:59:03 +08:00

top_k.py

[AMD] Dsv4/pr1 fix run time issue (#25898)

2026-05-23 16:04:14 -07:00

utils.py

[MUSA] Use MUSA-optimized operators in piecewise CUDA graph (#23633)

2026-05-11 17:55:27 -07:00

version.py

chore: bump sgl-kernel version to 0.4.4 (#28556)

2026-06-17 13:38:00 -07:00