Tao LiandXiaoyu Zhang c53cc8e1eb [NPU][BugFix] Avoid M-RoPE recompilation for variable sequence lengths (#40371) ...

Co-authored-by: Xiaoyu Zhang <1182563586@qq.com>

2026-09-22 08:19:25 +08:00

..

activation

[Refactor] Deduplicate kernel helpers and remove unused code (#40197)

2026-09-19 09:27:30 +08:00

attention

[NPU][BugFix] Avoid M-RoPE recompilation for variable sequence lengths (#40371)

2026-09-22 08:19:25 +08:00

communication

[DSV4.1] Reduce mHC, metadata and small-batch router overhead (#39704)

2026-09-19 19:51:21 +08:00

diffusion

[diffusion] CI: guard E2E/loading latency with runner-aware baselines (#39206)

2026-09-21 08:43:22 +08:00

elementwise

support qwen 3.8 flash next (#37500)

2026-09-08 13:56:21 -07:00

embeddings

dsv4.1: standalone kernels and Python wrappers (#39646)

2026-09-15 22:31:18 -07:00

gemm

[Refactor] Deduplicate kernel helpers and remove unused code (#40197)

2026-09-19 09:27:30 +08:00

grammar

[CI][RFC] Replace black-jupyter with ruff-format (#37210)

2026-09-02 19:46:08 -07:00

kimi_k3

[Refactor] Deduplicate kernel helpers and remove unused code (#40197)

2026-09-19 09:27:30 +08:00

kv_canary

Fix KV-canary workspace accounting after graph capture (#38596)

2026-09-10 16:10:12 -07:00

kvcache

[HiCache] TMA-staged host<->device KV transfer kernel (sm_90+) (#40278)

2026-09-21 10:38:23 -07:00

layernorm

[DSV4.1] Reduce mHC, metadata and small-batch router overhead (#39704)

2026-09-19 19:51:21 +08:00

lplb

[CI][RFC] Replace black-jupyter with ruff-format (#37210)

2026-09-02 19:46:08 -07:00

mamba

[XPU][Fix] Pack device-pointer tables as uint64 to avoid 64-bit address overflow (#35051)

2026-09-10 10:12:18 +08:00

memory

[Perf] Fuse SWA page lookup and mapping clear (#38948)

2026-09-18 15:51:07 -07:00

minicpm_sala

Refactor JIT kernel and expert-pack directory layout (#36704)

2026-08-29 07:41:25 +08:00

mm

feat(kernels): port standalone Kimi K3 kernels (#32890)

2026-08-01 13:26:47 +08:00

moe

[DSV4.1] Reduce mHC, metadata and small-batch router overhead (#39704)

2026-09-19 19:51:21 +08:00

quantization

[mxfp8-kv] Skip writes to the reserved CUDA-graph padding slot (#35351)

2026-09-22 01:21:25 +08:00

sampling

[CPU] Add native CPU kernel for MurmurHash32 (#35604)

2026-09-07 09:10:35 +08:00

speculative

[Spec] Windowed draft-decode attention for built-in EAGLE / MTP drafts (#32673)

2026-09-22 08:17:57 +08:00

__init__.py

Refactor JIT kernel and expert-pack directory layout (#36704)

2026-08-29 07:41:25 +08:00

qwen4_ple.py

support qwen 3.8 flash next (#37500)

2026-09-08 13:56:21 -07:00