[Score API][18132] return token usage in Score API response (#18381)

test/registered/openai_server/basic/test_openai_server.py

@@ -1038,6 +1038,18 @@ class TestOpenAIV1Score(CustomTestCase):
                 msg=f"Score {i} probabilities should sum to 1",
             )
 
+        # Verify usage
+        self.assertIn("usage", response, "Response should have a 'usage' field")
+        self.assertGreater(response["usage"]["prompt_tokens"], 0)
+        self.assertEqual(
+            response["usage"]["prompt_tokens"], response["usage"]["total_tokens"]
+        )
+        self.assertEqual(
+            response["usage"]["completion_tokens"],
+            0,
+            "completion_tokens should be 0 for /v1/score",
+        )
+
     def test_score_token_input(self):
         """Test scoring with token IDs input"""
         query = "The capital of France is"
@@ -1088,6 +1100,18 @@ class TestOpenAIV1Score(CustomTestCase):
                 msg=f"Score {i} probabilities should sum to 1",
             )
 
+        # Verify usage
+        self.assertIn("usage", response, "Response should have a 'usage' field")
+        self.assertGreater(response["usage"]["prompt_tokens"], 0)
+        self.assertEqual(
+            response["usage"]["prompt_tokens"], response["usage"]["total_tokens"]
+        )
+        self.assertEqual(
+            response["usage"]["completion_tokens"],
+            0,
+            "completion_tokens should be 0 for /v1/score",
+        )
+
     def test_score_error_handling(self):
         """Test error handling for invalid inputs"""
         query = "The capital of France is"

test/registered/openai_server/basic/test_serving_rerank.py

@@ -3,6 +3,7 @@ import unittest
 from unittest.mock import Mock
 
 from sglang.srt.entrypoints.openai.protocol import V1RerankReqInput
+from sglang.srt.managers.tokenizer_manager_multiitem_mixin import ScoreResult
 from sglang.test.ci.ci_register import register_amd_ci, register_cuda_ci
 
 # Keep consistent with other openai_server/basic unit tests.
@@ -163,7 +164,7 @@ class TestOpenAIServingRerankUnit(unittest.TestCase):
                 # Return [p_yes, p_no] for each prompt
                 assert len(prompts) == 2
                 assert label_token_ids and len(label_token_ids) == 2
-                return [[0.9, 0.1], [0.2, 0.8]]
+                return ScoreResult(scores=[[0.9, 0.1], [0.2, 0.8]], prompt_tokens=42)
 
         handler = OpenAIServingRerank(_TM())
         req = V1RerankReqInput(query="q", documents=["d1", "d2"], return_documents=True)