diff --git a/tests/entrypoints/openai/test_truncation.py b/tests/entrypoints/openai/test_truncation.py
index 18ddc493c9283..121c0413e1af7 100644
--- a/tests/entrypoints/openai/test_truncation.py
+++ b/tests/entrypoints/openai/test_truncation.py
@@ -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