Wentao Ye 837eae6458 [DSv4 Perf] Remove redundant full kernel for dsv4, 1.88x kernel performance improvement (#50298) ...

Signed-off-by: yewentao256 <zhyanwentao@126.com>

2026-07-30 11:39:07 -04:00

..

ops

[CompressedTensors] DeepSeek4 CT Quantization Support (#41276)

2026-07-23 20:07:31 -07:00

__init__.py

[Model Refactoring] Migrate DeepSeek V4 to vllm/models/ [1/N] (#43004)

2026-05-18 19:50:02 -07:00

dspark.py

[Bugfix] Fix DeepSeek-V4 DSpark draft shared-expert padding for TP > 8 (#49415)

2026-07-23 05:06:21 +00:00

flashinfer_sparse.py

[Bugfix] DSV4 TP16 garbage output (#47493)

2026-07-07 21:04:33 -07:00

flashmla.py

[DSv4 Perf] Remove redundant full kernel for dsv4, 1.88x kernel performance improvement (#50298)

2026-07-30 11:39:07 -04:00

model.py

[DSv4 Perf] Fix redundant memory allocation and copy for dsv4 pp buffer, 448 MiB GPU memory saved (#50312)

2026-07-30 11:38:51 -04:00

mtp.py

[Bugfix] Fix DeepSeek-V4 DSpark draft shared-expert padding for TP > 8 (#49415)

2026-07-23 05:06:21 +00:00