xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-17 06:55:01 +08:00
Code Issues Packages Projects Releases Wiki Activity

Fix test_memory_usage_no_spec (#17754)

Browse Source 
Signed-off-by: Yong Hoon Shin <yhshin@meta.com>
This commit is contained in:
Yong Hoon Shin 2025-05-07 00:10:33 -07:00 committed by GitHub
parent 8a15c2603a
commit 324a3119b0
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 6 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
tests/spec_decode/test_memory_usage.py
View File

@ -42,12 +42,12 @@ we can ensure we go through the _no_spec codepath for most of our engine steps.
def test_memory_usage_no_spec(): def test_memory_usage_no_spec():
previous_memory_allocated = None previous_memory_allocated = None
llm = vllm.LLM( llm = vllm.LLM(model=MAIN_MODEL,
model=MAIN_MODEL, speculative_config={
speculative_model=SPEC_MODEL, "model": SPEC_MODEL,
num_speculative_tokens=3, "num_speculative_tokens": 3,
speculative_disable_by_batch_size=SPEC_DISABLE_BATCH_SIZE, "disable_by_batch_size": SPEC_DISABLE_BATCH_SIZE,
) })
batch_sequences = set() batch_sequences = set()
engine = llm.llm_engine engine = llm.llm_engine