Commit Graph

2 Commits

Author SHA1 Message Date
Tony
b1557abab6 fix multistep bug,remove uselesscodes (#355)
1. remove useluss code in attention.py
2. multistep now using StatefulModelInputForNPU and do not use
StatefulModelInput

Signed-off-by: new-TonyWang <wangtonyyu222@gmail.com>
2025-03-28 09:55:35 +08:00
Tony
4c9d78a035 support multistep decode (#299)
Add multi step scheduler support for vllm-ascend

Signed-off-by: new-TonyWang <wangtonyyu222@gmail.com>
2025-03-11 19:20:06 +08:00