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

Improve enable chunked_prefill & prefix_caching logic. (#26623)

Browse Source 
Signed-off-by: wang.yuqi <yuqi.wang@daocloud.io>
Signed-off-by: wang.yuqi <noooop@126.com>
Co-authored-by: Cyrus Leung <tlleungac@connect.ust.hk>
This commit is contained in:
wang.yuqi
2025-11-28 14:05:48 +08:00
committed by GitHub
parent 37b15e97e8
commit f4b76056ee
11 changed files with 456 additions and 133 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
tests/models/language/pooling/test_auto_prefix_cache_support.py
View File

@@ -105,8 +105,6 @@ def test_embed_models(
def test_non_causal_models(
hf_runner, vllm_runner, example_prompts, model: str, dtype: str
) -> None:
with vllm_runner(
model, max_model_len=512, dtype=dtype, enable_prefix_caching=True
) as vllm_model:
with vllm_runner(model, max_model_len=512, dtype=dtype) as vllm_model:
cache_config = vllm_model.llm.llm_engine.cache_config
assert not cache_config.enable_prefix_caching