whx ee2ed573f1 [BugFix][DS 3.2] Fix ds indexer accuracy problem caused by rope. (#4641) ...

### What this PR does / why we need it?
The rotary algorithm in deepseek indexer should be neox-style instead of
gptj style. PR #4413 fix this accuracy bug with new triton kernel. This
PR fixes original pytorch version.

### Does this PR introduce _any_ user-facing change?
None

### How was this patch tested?
CI passed with existing test.


- vLLM version: 86e178f7c4d8c3b0eaf3c8e3f810a83f63b90e24
- vLLM main:
86e178f7c4

Signed-off-by: whx-sjtu <2952154980@qq.com>

2026-01-09 14:11:44 +08:00

..

context_parallel

[feature]dcp&pcp support mlapo (#5672)

2026-01-08 23:49:23 +08:00

__init__.py

[Core] Make V1 work and enable V1 engine test (#389)

2025-03-28 19:34:23 +08:00

attention_mask.py

[Refactor] Fix AttentionMaskBuilder singleton and remove redundant pcp_prefill_mask (#4870)

2026-01-07 17:09:52 +08:00

attention_v1.py

[Refactor] Fix AttentionMaskBuilder singleton and remove redundant pcp_prefill_mask (#4870)

2026-01-07 17:09:52 +08:00

mla_v1.py

[feature]dcp&pcp support mlapo (#5672)

2026-01-08 23:49:23 +08:00

sfa_v1.py

[BugFix][DS 3.2] Fix ds indexer accuracy problem caused by rope. (#4641)

2026-01-09 14:11:44 +08:00

utils.py

[Fix] Fixes speculative decode indexing and unpad condition for attention metadata (#5626)

2026-01-08 19:41:08 +08:00