diff --git a/tests/entrypoints/openai/test_completion_error.py b/tests/entrypoints/openai/test_completion_error.py
index a7f6a75e0e72..e48cc32e5400 100644
--- a/tests/entrypoints/openai/test_completion_error.py
+++ b/tests/entrypoints/openai/test_completion_error.py
@@ -219,3 +219,23 @@ async def mock_generate(*args, **kwargs):
         f"Expected error message in chunks: {chunks}"
     )
     assert chunks[-1] == "data: [DONE]\n\n"
+
+
+def test_negative_prompt_token_ids_nested():
+    """Negative token IDs in prompt (nested list) should raise validation error."""
+    with pytest.raises(Exception, match="greater than or equal to 0"):
+        CompletionRequest(
+            model=MODEL_NAME,
+            prompt=[[-1]],
+            max_tokens=10,
+        )
+
+
+def test_negative_prompt_token_ids_flat():
+    """Negative token IDs in prompt (flat list) should raise validation error."""
+    with pytest.raises(Exception, match="greater than or equal to 0"):
+        CompletionRequest(
+            model=MODEL_NAME,
+            prompt=[-1],
+            max_tokens=10,
+        )
diff --git a/vllm/entrypoints/openai/completion/protocol.py b/vllm/entrypoints/openai/completion/protocol.py
index aec1a0a95de6..531de984b92b 100644
--- a/vllm/entrypoints/openai/completion/protocol.py
+++ b/vllm/entrypoints/openai/completion/protocol.py
@@ -42,7 +42,13 @@ class CompletionRequest(OpenAIBaseModel):
     # Ordered by official OpenAI API documentation
     # https://platform.openai.com/docs/api-reference/completions/create
     model: str | None = None
-    prompt: list[int] | list[list[int]] | str | list[str] | None = None
+    prompt: (
+        list[Annotated[int, Field(ge=0)]]
+        | list[list[Annotated[int, Field(ge=0)]]]
+        | str
+        | list[str]
+        | None
+    ) = None
     echo: bool | None = False
     frequency_penalty: float | None = 0.0
     logit_bias: dict[str, float] | None = None