mirror of
https://git.datalinker.icu/vllm-project/vllm.git
synced 2025-12-14 00:34:58 +08:00
68 lines
2.3 KiB
Python
68 lines
2.3 KiB
Python
# SPDX-License-Identifier: Apache-2.0
|
|
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
|
|
|
|
from collections.abc import Sequence
|
|
|
|
from vllm.entrypoints.openai.protocol import DeltaMessage
|
|
from vllm.reasoning.basic_parsers import BaseThinkingReasoningParser
|
|
|
|
|
|
class DeepSeekR1ReasoningParser(BaseThinkingReasoningParser):
|
|
"""
|
|
Reasoning parser for DeepSeek R1 model.
|
|
|
|
The DeepSeek R1 model uses <think>...</think> tokens to denote reasoning
|
|
text. This parser extracts the reasoning content from the model output.
|
|
"""
|
|
|
|
@property
|
|
def start_token(self) -> str:
|
|
"""The token that starts reasoning content."""
|
|
return "<think>"
|
|
|
|
@property
|
|
def end_token(self) -> str:
|
|
"""The token that ends reasoning content."""
|
|
return "</think>"
|
|
|
|
def extract_reasoning_streaming(
|
|
self,
|
|
previous_text: str,
|
|
current_text: str,
|
|
delta_text: str,
|
|
previous_token_ids: Sequence[int],
|
|
current_token_ids: Sequence[int],
|
|
delta_token_ids: Sequence[int],
|
|
) -> DeltaMessage | None:
|
|
ret = super().extract_reasoning_streaming(
|
|
previous_text,
|
|
current_text,
|
|
delta_text,
|
|
previous_token_ids,
|
|
current_token_ids,
|
|
delta_token_ids,
|
|
)
|
|
if (
|
|
ret is not None
|
|
and self.start_token_id not in previous_token_ids
|
|
and self.start_token_id not in delta_token_ids
|
|
):
|
|
if self.end_token_id in delta_token_ids:
|
|
# end token in delta with more tokens,
|
|
# extract reasoning content and content
|
|
end_index = delta_text.find(self.end_token)
|
|
reasoning = delta_text[:end_index]
|
|
content = delta_text[end_index + len(self.end_token) :]
|
|
return DeltaMessage(
|
|
reasoning=reasoning,
|
|
content=content if content else None,
|
|
)
|
|
elif self.end_token_id in previous_token_ids:
|
|
# end token in previous, thinking content ends
|
|
return DeltaMessage(content=delta_text)
|
|
else:
|
|
# no end token in previous or delta, reasoning content continues
|
|
return DeltaMessage(reasoning=delta_text)
|
|
|
|
return ret
|