[Core] Adding token ranks along with logprobs (#3516)

Co-authored-by: Swapnil Parekh <swapnilp@ibm.com>

vllm/sequence.py

@@ -16,8 +16,15 @@ if TYPE_CHECKING:
 
 @dataclass
 class Logprob:
-    """Infos for supporting OpenAI compatible logprobs."""
+    """Infos for supporting OpenAI compatible logprobs and token ranks.
+
+    Attributes:
+        logprob: The logprob of chosen token
+        rank: The vocab rank of chosen token (>=1)
+        decoded_token: The decoded chosen token index
+    """
     logprob: float
+    rank: Optional[int] = None
     decoded_token: Optional[str] = None
 
 
@@ -66,7 +73,7 @@ class SequenceStatus(enum.Enum):
 class RequestMetrics:
     """Metrics associated with a request.
 
-    Args:
+    Attributes:
         arrival_time: The time when the request arrived.
         first_scheduled_time: The time when the request was first scheduled.
         first_token_time: The time when the first token was generated.