[Refactor] Consolidate sequence normalization and enc-dec parsing (#33928)

Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
This commit is contained in:
Cyrus Leung
2026-02-06 15:43:47 +00:00
committed by GitHub
parent 4707f7ebb4
commit cd8b405bd0
38 changed files with 1271 additions and 863 deletions
+4 -3
View File
@@ -22,6 +22,8 @@ from vllm.outputs import PoolingRequestOutput, RequestOutput
from vllm.plugins.io_processors import get_io_processor
from vllm.pooling_params import PoolingParams
from vllm.renderers import BaseRenderer
from vllm.renderers.inputs import DictPrompt, TokPrompt
from vllm.renderers.inputs.preprocess import extract_prompt_components
from vllm.sampling_params import SamplingParams
from vllm.tasks import SupportedTask
from vllm.tokenizers import TokenizerLike
@@ -32,7 +34,6 @@ from vllm.v1.engine.core_client import EngineCoreClient
from vllm.v1.engine.input_processor import InputProcessor
from vllm.v1.engine.output_processor import OutputProcessor
from vllm.v1.engine.parallel_sampling import ParentRequest
from vllm.v1.engine.utils import get_prompt_text
from vllm.v1.executor import Executor
from vllm.v1.metrics.loggers import StatLoggerFactory, StatLoggerManager
from vllm.v1.metrics.reader import Metric, get_metrics_snapshot
@@ -216,7 +217,7 @@ class LLMEngine:
def add_request(
self,
request_id: str,
prompt: EngineCoreRequest | PromptType,
prompt: EngineCoreRequest | PromptType | DictPrompt | TokPrompt,
params: SamplingParams | PoolingParams,
arrival_time: float | None = None,
lora_request: LoRARequest | None = None,
@@ -251,7 +252,7 @@ class LLMEngine:
priority,
supported_tasks=self.get_supported_tasks(),
)
prompt_text = get_prompt_text(prompt)
prompt_text, _, _ = extract_prompt_components(self.model_config, prompt)
self.input_processor.assign_request_id(request)