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

[V0 Deprecation] Remove pooling model support in V0 (#23434)

Browse Source 
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
Signed-off-by: Max de Bayser <mbayser@br.ibm.com>
Co-authored-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
This commit is contained in:
Maximilien de Bayser
2025-08-29 04:04:02 -03:00
committed by GitHub
parent 934bebf192
commit 2554b27baa
38 changed files with 99 additions and 808 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
tests/distributed/test_pipeline_parallel.py
View File

@@ -118,6 +118,8 @@ class PPTestSettings:
multi_node_only: bool = False,
load_format: Optional[str] = None,
):
vllm_major_versions = ["1"] if runner == "pooling" else ["0"]
return PPTestSettings(
parallel_setups=[
ParallelSetup(tp_size=tp_base,
@@ -126,7 +128,7 @@ class PPTestSettings:
chunked_prefill=False),
],
distributed_backends=["mp"],
vllm_major_versions=["0"],
vllm_major_versions=vllm_major_versions,
runner=runner,
test_options=PPTestOptions(multi_node_only=multi_node_only,
load_format=load_format),
@@ -213,7 +215,9 @@ TEXT_GENERATION_MODELS = {
EMBEDDING_MODELS = { # type: ignore[var-annotated]
# [Text-only]
"intfloat/e5-mistral-7b-instruct": PPTestSettings.fast(runner="pooling"),
"BAAI/bge-multilingual-gemma2": PPTestSettings.fast(runner="pooling"),
# TODO: re-enable when https://github.com/vllm-project/vllm/issues/23883
# is fixed
#"BAAI/bge-multilingual-gemma2": PPTestSettings.fast(runner="pooling"),
"Qwen/Qwen2.5-Math-RM-72B": PPTestSettings.fast(
load_format="dummy", runner="pooling"
),