biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[core][distributed] support n layers % pp size != 0 (#6115)

Browse Source
This commit is contained in:
youkaichao
2024-07-03 16:40:31 -07:00
committed by GitHub
parent 966fe72141
commit 3de6e6a30e
7 changed files with 19 additions and 10 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
.buildkite/test-pipeline.yaml
View File

@@ -80,6 +80,7 @@ steps:
commands:
- TP_SIZE=2 PP_SIZE=2 EAGER_MODE=1 CHUNKED_PREFILL=1 pytest -v -s distributed/test_pipeline_parallel.py
- TP_SIZE=2 PP_SIZE=2 EAGER_MODE=1 CHUNKED_PREFILL=0 pytest -v -s distributed/test_pipeline_parallel.py
- TP_SIZE=1 PP_SIZE=3 EAGER_MODE=1 CHUNKED_PREFILL=0 pytest -v -s distributed/test_pipeline_parallel.py
- PP_SIZE=4 EAGER_MODE=1 CHUNKED_PREFILL=1 pytest -v -s distributed/test_pipeline_parallel.py
- PP_SIZE=4 EAGER_MODE=1 CHUNKED_PREFILL=0 pytest -v -s distributed/test_pipeline_parallel.py