[V1] Logprobs and prompt logprobs support (#9880)

This PR is adding support for sample logprobs & prompt logprobs to vLLM v1.

New behavior:

- During model execution, model runner computes sample logprobs (if user-provided logprobs setting is not None) and prompt logprobs (if user-provided prompt_logprobs setting is not None). For both sample and prompt logprobs, the engine core returns 3 vectors: token ids, token logprob values, token ranks. Ranks reflect tokens' 1-indexed positions in the vocabulary vector after sorting the vocabulary by log probability in descending order.
- In scheduler.update_from_output(), sample and prompt logprobs are incorporated into the EngineCoreOutput data structure which is transferred to the engine client. If multiprocessing is enabled, then sample and prompt logprobs will be (de)serialized when the EngineCoreOutput data structure is (de)serialized.
- During output processing, the LogprobsProcessor transforms the triplet of token ids, token logprobs values, and token ranks into the OpenAI-compatible List[Dict[token id,Logprob]] format (for sample and prompt logprobs respectively.)
- Each Logprob instance (whether sample- or prompt-) consists of a token's log-probability, rank, and detokenized string representation. Note that logprob detokenization is handled by the LogprobsProcessor not the detokenizer.

Signed-off-by: Andrew Feldman <afeldman@neuralmagic.com>
Signed-off-by: Nick Hill <nhill@redhat.com>
Signed-off-by: rshaw@neuralmagic.com <rshaw@neuralmagic.com>


Co-authored-by: rshaw@neuralmagic.com <rshaw@neuralmagic.com>
Co-authored-by: Nick Hill <nhill@redhat.com>

vllm/v1/engine/core.py

@@ -11,7 +11,6 @@ from typing import List, Tuple, Type
 import psutil
 import zmq
 import zmq.asyncio
-from msgspec import msgpack
 
 from vllm.config import VllmConfig
 from vllm.logger import init_logger
@@ -26,7 +25,7 @@ from vllm.v1.engine import (EngineCoreOutputs, EngineCoreProfile,
 from vllm.v1.engine.mm_input_mapper import MMInputMapperServer
 from vllm.v1.executor.abstract import Executor
 from vllm.v1.request import Request, RequestStatus
-from vllm.v1.serial_utils import PickleEncoder
+from vllm.v1.serial_utils import MsgpackEncoder, PickleEncoder
 from vllm.version import __version__ as VLLM_VERSION
 
 logger = init_logger(__name__)
@@ -292,7 +291,7 @@ class EngineCoreProc(EngineCore):
         """Output socket IO thread."""
 
         # Msgpack serialization encoding.
-        encoder = msgpack.Encoder()
+        encoder = MsgpackEncoder()
         # Reuse send buffer.
         buffer = bytearray()