xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-10 03:54:56 +08:00
Code Issues Packages Projects Releases Wiki Activity

(Misc): add missing test for zero truncation size. (#23457)

Browse Source 
Signed-off-by: teekenl <teekenlau@gmail.com>
This commit is contained in:
TeeKen Lau 2025-08-24 20:12:47 +10:00 committed by GitHub
parent 1b9b16649c
commit 5e021b4981
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 22 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

22
tests/entrypoints/openai/test_truncation.py
View File

@ -64,6 +64,28 @@ async def test_smaller_truncation_size(client: openai.AsyncOpenAI):
assert response["usage"]["prompt_tokens"] == truncation_size
@pytest.mark.asyncio
async def test_zero_truncation_size(client: openai.AsyncOpenAI):
truncation_size = 0
kwargs: dict[str, Any] = {
"model": MODEL_NAME,
"input": input,
"truncate_prompt_tokens": truncation_size
}
with pytest.raises(openai.BadRequestError) as err:
await client.post(path="embeddings", cast_to=object, body={**kwargs})
assert err.value.status_code == 400
error_details = err.value.response.json()["error"]
assert error_details["type"] == "BadRequestError"
assert "This model's maximum context length is" in error_details["message"]
assert "tokens in the input for embedding generation" in error_details[
"message"]
assert "Please reduce the length of the input" in error_details["message"]
@pytest.mark.asyncio
async def test_bigger_truncation_size(client: openai.AsyncOpenAI):
truncation_size = max_model_len + 1