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

Support encoder-only models without KV-Cache (#21270)

Browse Source 
Signed-off-by: Max de Bayser <maxdebayser@gmail.com>
Signed-off-by: Max de Bayser <mbayser@br.ibm.com>
Co-authored-by: Russell Bryant <rbryant@redhat.com>
This commit is contained in:
Maximilien de Bayser
2025-07-26 10:09:52 -03:00
committed by GitHub
parent f27fdfc3ed
commit 1cd6eaba54
17 changed files with 352 additions and 99 deletions
Download Patch File Download Diff File Expand all files Collapse all files

14
tests/models/language/pooling/test_embedding.py
View File

@@ -39,17 +39,9 @@ def v1(run_with_both_engines):
pytest.param("ssmits/Qwen2-7B-Instruct-embed-base",
marks=[pytest.mark.skip_v0, pytest.mark.cpu_model]),
# [Encoder-only]
pytest.param(
"BAAI/bge-base-en-v1.5",
marks=[
# CPU only supports V1
pytest.mark.core_model,
pytest.mark.skip_v1
]),
pytest.param("sentence-transformers/all-MiniLM-L12-v2",
marks=[pytest.mark.skip_v1]),
pytest.param("intfloat/multilingual-e5-small",
marks=[pytest.mark.skip_v1]),
pytest.param("BAAI/bge-base-en-v1.5", marks=[pytest.mark.core_model]),
pytest.param("sentence-transformers/all-MiniLM-L12-v2"),
pytest.param("intfloat/multilingual-e5-small"),
pytest.param("Alibaba-NLP/gte-Qwen2-1.5B-instruct",
marks=[pytest.mark.skip_v1]),
# [Cross-Encoder]