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

fix: use cache_salt for gpt-oss (#23186)

Browse Source 
Signed-off-by: Marko Rosenmueller <5467316+dr75@users.noreply.github.com>
This commit is contained in:
Marko Rosenmueller
2025-08-19 20:12:25 +02:00
committed by GitHub
parent b94faf9d50
commit 80141bbf2f
3 changed files with 13 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
tests/entrypoints/openai/test_serving_chat.py
View File

@@ -282,9 +282,11 @@ async def test_serving_chat_could_load_correct_generation_config():
assert mock_engine.generate.call_args.args[1].repetition_penalty == 1.05
@pytest.mark.parametrize("model_type", ["gpt_oss", "any"])
@pytest.mark.asyncio
async def test_serving_chat_did_set_correct_cache_salt():
async def test_serving_chat_did_set_correct_cache_salt(model_type):
mock_model_config = MockModelConfig()
mock_model_config.hf_config.model_type = model_type
mock_engine = MagicMock(spec=MQLLMEngineClient)
mock_engine.get_tokenizer.return_value = get_tokenizer(MODEL_NAME)