Khoa PhamandQiaolin Yu 018b73c7a0 [PD] Pack draft KV head slices for DCP transfers (#40500) ...

Co-authored-by: Qiaolin Yu <liin1211@outlook.com>

2026-09-21 21:11:27 -07:00

..

activation

fix(moe): cast filtered-activation expert_ids to int32 for torch.compile (#38085)

2026-09-05 17:13:07 +08:00

attention

[AMD] [GLM-5.3-Flash Day 0] Support non-2048 top-k widths in the DSA page-table transform (#39340)

2026-09-21 19:50:17 -07:00

communication

[Refactor] Clean up parallel runtime comments (#40632)

2026-09-21 14:32:22 -07:00

cuda_graph

[Test] Consolidate kernel tests under plural kernels tree (#39966)

2026-09-18 07:37:48 +08:00

diffusion

[diffusion] model: support qwen-image-2.1 (#39983)

2026-09-20 09:46:09 +08:00

disaggregation

[Test] Consolidate kernel tests under plural kernels tree (#39966)

2026-09-18 07:37:48 +08:00

elementwise

[Test] Consolidate kernel tests under plural kernels tree (#39966)

2026-09-18 07:37:48 +08:00

embeddings

[Qwen4-Exp] Build the offloaded PLE table on the meta device so --ple-offload-embedding never materialises it on the accelerator (#39928)

2026-09-20 08:28:25 -07:00

gemm

[CI] Fix stale ServerArgs fake in chunked-SGMV LoRA test (#38315)

2026-09-07 04:25:12 -07:00

hyperconnection

[Test] Consolidate kernel tests under plural kernels tree (#39966)

2026-09-18 07:37:48 +08:00

kimi_k3

[Refactor] Clean up parallel runtime comments (#40632)

2026-09-21 14:32:22 -07:00

kv_canary

[Intel][XPU][KVCanary] Enable KV Canary on Intel XPU (#33520)

2026-09-22 09:19:01 +08:00

kvcache

[PD] Pack draft KV head slices for DCP transfers (#40500)

2026-09-21 21:11:27 -07:00

layernorm

[Refactor] Clean up parallel runtime comments (#40632)

2026-09-21 14:32:22 -07:00

mamba

[CI] Give the kernel lane a 5090 suite and move kernel-only tests off the general lane (#40496)

2026-09-20 16:46:53 -07:00

memory

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

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

minicpm_sala

[CI] Derive registered-test kind from the registry call instead of the path (#40294)

2026-09-20 02:01:22 -07:00

moe

[AMD] [GLM-5.3-Flash Day 0] Build the fused DSA k-pool top-k JIT kernel on HIP (#39339)

2026-09-21 15:24:44 -07:00

qsa

[Test] Consolidate kernel tests under plural kernels tree (#39966)

2026-09-18 07:37:48 +08:00

quantization

[CI] Derive registered-test kind from the registry call instead of the path (#40294)

2026-09-20 02:01:22 -07:00

sampling

perf(sampling): avoid GPU syncs when applying custom logit processors (#39234)

2026-09-18 17:09:30 -07:00

speculative

Fix lint failure from draft-decode window test location (#40654)

2026-09-21 17:41:00 -07:00