Ryan Zzzandzhujunyu 8fd1694dd2 Deepseek v4: support mixed dtype compression states (#27277) ...

Co-authored-by: zhujunyu <zhujunyu.666@bytedance.com>

2026-06-17 00:56:41 -07:00

..

__init__.py

[PD & HiSparse] Add DeepSeek V4 support for HiSparse direct Prefill-to-Decode DRAM (#24880)

2026-06-05 15:39:48 +08:00

attn.py

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

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

compress_old.py

[Refactor] major JIT kernel clean up for dsv4 (#25884)

2026-05-21 01:14:31 -07:00

compress.py

Deepseek v4: support mixed dtype compression states (#27277)

2026-06-17 00:56:41 -07:00

elementwise.py

Add FP4 Indexer for DeepSeek V4 (#26209)

2026-06-02 00:14:38 -07:00

gemm.py

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

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

moe.py

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

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

online_c128_mtp.py

DeepSeek-V4 Online Compress support MTP (#26471)

2026-06-15 19:56:07 -07:00

topk.py

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

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

utils.py

Port KV Compression V2 from deepseek_v4_dev (#24890)

2026-05-13 22:40:38 +08:00