apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2026-04-24 09:44:10 +08:00
Code Issues Actions 9 Packages Projects Releases Wiki Activity

[Feature] Support mtp overlap schedule (#7001)

Browse Source
This commit is contained in:
sunxin
2026-04-01 14:24:26 +08:00
committed by GitHub
parent c6f0c5c3a6
commit c29e86fc9d
23 changed files with 215 additions and 138 deletions
Download Patch File Download Diff File Expand all files Collapse all files
fastdeploy/model_executor/layers/sample/logprobs.py
+1 -1
View File
@@ -123,7 +123,7 @@ def gather_logprobs(
indices = token_ids
top_logprobs = token_logprobs
return LogprobsTensors(indices, top_logprobs, token_ranks)
return LogprobsTensors(indices.cpu(), top_logprobs.cpu(), token_ranks.cpu())
def build_output_logprobs(