freeliuzc 2d1dade5e2 [Speculative Decoding][MTP] Support static CacheKV C8 quantization and optimize memory usage (#5155) ...

* support static cachekv c8 quantization in mtp mode

* optimize memory allocation

2025-11-21 15:10:13 +08:00

..

__init__.py

…

base.py

[Speculative Decoding][MTP]Support attn mask offset (#4641)

2025-11-03 10:08:01 +08:00

mtp.py

[Speculative Decoding][MTP] Support static CacheKV C8 quantization and optimize memory usage (#5155)

2025-11-21 15:10:13 +08:00

ngram.py

…