xc-llm-ascend

Files

dsxsteven 847d12a389 [BugFix]Fix moe load problems in torchair when using dynamic eplb (#3381 )

### What this PR does / why we need it?

When using dynamic eplb, moe load is not imported. We fix this problem
by modifying the return value of hidden states in torchair.

### Does this PR introduce _any_ user-facing change?

### How was this patch tested?
DeepseekV3 in A3.


- vLLM version: v0.11.0rc3
- vLLM main: https://github.com/vllm-project/vllm/commit/v0.11.0

---------

Signed-off-by: daishixun <dsxsteven@sina.com>

2025-10-13 11:38:57 +08:00

models

[1/N][Feat] Add weight prefetch feature for Attention layers (#3146 )

2025-10-09 20:38:39 +08:00

ops

[BugFix]Fix moe load problems in torchair when using dynamic eplb (#3381 )

2025-10-13 11:38:57 +08:00

quantization

[BugFix]Fix moe load problems in torchair when using dynamic eplb (#3381 )

2025-10-13 11:38:57 +08:00

__init__.py

[1/4][Refactor] Refactor torchair worker (#1885 )