[V0 Deprecation] Remove Prompt Adapters (#20588)

Signed-off-by: mgoin <mgoin64@gmail.com>
This commit is contained in:
Michael Goin
2025-07-23 16:36:48 -07:00
committed by GitHub
parent 78c13e30e1
commit 82ec66f514
60 changed files with 126 additions and 1727 deletions
+2 -44
View File
@@ -45,7 +45,6 @@ from vllm.outputs import (ClassificationRequestOutput, EmbeddingRequestOutput,
PoolingRequestOutput, RequestOutput,
ScoringRequestOutput)
from vllm.pooling_params import PoolingParams, PoolingTask
from vllm.prompt_adapter.request import PromptAdapterRequest
from vllm.sampling_params import (BeamSearchParams, GuidedDecodingParams,
RequestOutputKind, SamplingParams)
from vllm.transformers_utils.tokenizer import (AnyTokenizer, MistralTokenizer,
@@ -314,7 +313,6 @@ class LLM:
*,
use_tqdm: Union[bool, Callable[..., tqdm]] = True,
lora_request: Optional[Union[list[LoRARequest], LoRARequest]] = None,
prompt_adapter_request: Optional[PromptAdapterRequest] = None,
guided_options_request: Optional[Union[LLMGuidedOptions,
GuidedDecodingRequest]] = None,
) -> list[RequestOutput]:
@@ -330,7 +328,6 @@ class LLM:
prompt_token_ids: Optional[list[int]] = None,
use_tqdm: Union[bool, Callable[..., tqdm]] = True,
lora_request: Optional[Union[list[LoRARequest], LoRARequest]] = None,
prompt_adapter_request: Optional[PromptAdapterRequest] = None,
guided_options_request: Optional[Union[LLMGuidedOptions,
GuidedDecodingRequest]] = None,
) -> list[RequestOutput]:
@@ -346,7 +343,6 @@ class LLM:
prompt_token_ids: Optional[list[list[int]]] = None,
use_tqdm: Union[bool, Callable[..., tqdm]] = True,
lora_request: Optional[Union[list[LoRARequest], LoRARequest]] = None,
prompt_adapter_request: Optional[PromptAdapterRequest] = None,
guided_options_request: Optional[Union[LLMGuidedOptions,
GuidedDecodingRequest]] = None,
) -> list[RequestOutput]:
@@ -363,7 +359,6 @@ class LLM:
prompt_token_ids: list[int],
use_tqdm: Union[bool, Callable[..., tqdm]] = True,
lora_request: Optional[Union[list[LoRARequest], LoRARequest]] = None,
prompt_adapter_request: Optional[PromptAdapterRequest] = None,
guided_options_request: Optional[Union[LLMGuidedOptions,
GuidedDecodingRequest]] = None,
) -> list[RequestOutput]:
@@ -380,7 +375,6 @@ class LLM:
prompt_token_ids: list[list[int]],
use_tqdm: Union[bool, Callable[..., tqdm]] = True,
lora_request: Optional[Union[list[LoRARequest], LoRARequest]] = None,
prompt_adapter_request: Optional[PromptAdapterRequest] = None,
guided_options_request: Optional[Union[LLMGuidedOptions,
GuidedDecodingRequest]] = None,
) -> list[RequestOutput]:
@@ -395,7 +389,6 @@ class LLM:
prompt_token_ids: Union[list[int], list[list[int]]],
use_tqdm: Union[bool, Callable[..., tqdm]] = True,
lora_request: Optional[Union[list[LoRARequest], LoRARequest]] = None,
prompt_adapter_request: Optional[PromptAdapterRequest] = None,
guided_options_request: Optional[Union[LLMGuidedOptions,
GuidedDecodingRequest]] = None,
) -> list[RequestOutput]:
@@ -415,7 +408,6 @@ class LLM:
prompt_token_ids: Optional[Union[list[int], list[list[int]]]] = None,
use_tqdm: Union[bool, Callable[..., tqdm]] = True,
lora_request: Optional[Union[list[LoRARequest], LoRARequest]] = None,
prompt_adapter_request: Optional[PromptAdapterRequest] = None,
guided_options_request: Optional[Union[LLMGuidedOptions,
GuidedDecodingRequest]] = None,
priority: Optional[list[int]] = None,
@@ -440,8 +432,6 @@ class LLM:
it is used to create the progress bar.
If `False`, no progress bar is created.
lora_request: LoRA request to use for generation, if any.
prompt_adapter_request: Prompt Adapter request to use for
generation, if any.
priority: The priority of the requests, if any.
Only applicable when priority scheduling policy is enabled.
@@ -507,7 +497,6 @@ class LLM:
params=sampling_params,
use_tqdm=use_tqdm,
lora_request=lora_request,
prompt_adapter_request=prompt_adapter_request,
guided_options=guided_options_request,
tokenization_kwargs=tokenization_kwargs,
priority=priority,
@@ -963,7 +952,6 @@ class LLM:
truncate_prompt_tokens: Optional[int] = None,
use_tqdm: Union[bool, Callable[..., tqdm]] = True,
lora_request: Optional[Union[list[LoRARequest], LoRARequest]] = None,
prompt_adapter_request: Optional[PromptAdapterRequest] = None,
pooling_task: PoolingTask = "encode",
tokenization_kwargs: Optional[dict[str, Any]] = None,
) -> list[PoolingRequestOutput]:
@@ -980,7 +968,6 @@ class LLM:
truncate_prompt_tokens: Optional[int] = None,
use_tqdm: Union[bool, Callable[..., tqdm]] = True,
lora_request: Optional[Union[list[LoRARequest], LoRARequest]] = None,
prompt_adapter_request: Optional[PromptAdapterRequest] = None,
pooling_task: PoolingTask = "encode",
tokenization_kwargs: Optional[dict[str, Any]] = None,
) -> list[PoolingRequestOutput]:
@@ -997,7 +984,6 @@ class LLM:
truncate_prompt_tokens: Optional[int] = None,
use_tqdm: Union[bool, Callable[..., tqdm]] = True,
lora_request: Optional[Union[list[LoRARequest], LoRARequest]] = None,
prompt_adapter_request: Optional[PromptAdapterRequest] = None,
pooling_task: PoolingTask = "encode",
tokenization_kwargs: Optional[dict[str, Any]] = None,
) -> list[PoolingRequestOutput]:
@@ -1015,7 +1001,6 @@ class LLM:
truncate_prompt_tokens: Optional[int] = None,
use_tqdm: Union[bool, Callable[..., tqdm]] = True,
lora_request: Optional[Union[list[LoRARequest], LoRARequest]] = None,
prompt_adapter_request: Optional[PromptAdapterRequest] = None,
pooling_task: PoolingTask = "encode",
tokenization_kwargs: Optional[dict[str, Any]] = None,
) -> list[PoolingRequestOutput]:
@@ -1033,7 +1018,6 @@ class LLM:
truncate_prompt_tokens: Optional[int] = None,
use_tqdm: Union[bool, Callable[..., tqdm]] = True,
lora_request: Optional[Union[list[LoRARequest], LoRARequest]] = None,
prompt_adapter_request: Optional[PromptAdapterRequest] = None,
pooling_task: PoolingTask = "encode",
tokenization_kwargs: Optional[dict[str, Any]] = None,
) -> list[PoolingRequestOutput]:
@@ -1049,7 +1033,6 @@ class LLM:
truncate_prompt_tokens: Optional[int] = None,
use_tqdm: Union[bool, Callable[..., tqdm]] = True,
lora_request: Optional[Union[list[LoRARequest], LoRARequest]] = None,
prompt_adapter_request: Optional[PromptAdapterRequest] = None,
pooling_task: PoolingTask = "encode",
tokenization_kwargs: Optional[dict[str, Any]] = None,
) -> list[PoolingRequestOutput]:
@@ -1070,7 +1053,6 @@ class LLM:
truncate_prompt_tokens: Optional[int] = None,
use_tqdm: Union[bool, Callable[..., tqdm]] = True,
lora_request: Optional[Union[list[LoRARequest], LoRARequest]] = None,
prompt_adapter_request: Optional[PromptAdapterRequest] = None,
pooling_task: PoolingTask = "encode",
tokenization_kwargs: Optional[dict[str, Any]] = None,
) -> list[PoolingRequestOutput]:
@@ -1092,8 +1074,6 @@ class LLM:
it is used to create the progress bar.
If `False`, no progress bar is created.
lora_request: LoRA request to use for generation, if any.
prompt_adapter_request: Prompt Adapter request to use for
generation, if any.
pooling_task: Override the pooling task to use.
Returns:
@@ -1150,7 +1130,6 @@ class LLM:
use_tqdm=use_tqdm,
lora_request=lora_request,
tokenization_kwargs=tokenization_kwargs,
prompt_adapter_request=prompt_adapter_request,
)
outputs = self._run_engine(use_tqdm=use_tqdm)
@@ -1167,7 +1146,6 @@ class LLM:
pooling_params: Optional[Union[PoolingParams,
Sequence[PoolingParams]]] = None,
lora_request: Optional[Union[list[LoRARequest], LoRARequest]] = None,
prompt_adapter_request: Optional[PromptAdapterRequest] = None,
) -> list[EmbeddingRequestOutput]:
"""
Generate an embedding vector for each prompt.
@@ -1187,8 +1165,6 @@ class LLM:
it is used to create the progress bar.
If `False`, no progress bar is created.
lora_request: LoRA request to use for generation, if any.
prompt_adapter_request: Prompt Adapter request to use for
generation, if any.
Returns:
A list of `EmbeddingRequestOutput` objects containing the
@@ -1205,7 +1181,6 @@ class LLM:
use_tqdm=use_tqdm,
pooling_params=pooling_params,
lora_request=lora_request,
prompt_adapter_request=prompt_adapter_request,
pooling_task="embed",
)
@@ -1218,7 +1193,6 @@ class LLM:
*,
use_tqdm: Union[bool, Callable[..., tqdm]] = True,
lora_request: Optional[Union[list[LoRARequest], LoRARequest]] = None,
prompt_adapter_request: Optional[PromptAdapterRequest] = None,
) -> list[ClassificationRequestOutput]:
"""
Generate class logits for each prompt.
@@ -1236,8 +1210,6 @@ class LLM:
it is used to create the progress bar.
If `False`, no progress bar is created.
lora_request: LoRA request to use for generation, if any.
prompt_adapter_request: Prompt Adapter request to use for
generation, if any.
Returns:
A list of `ClassificationRequestOutput` objects containing the
@@ -1253,7 +1225,6 @@ class LLM:
prompts,
use_tqdm=use_tqdm,
lora_request=lora_request,
prompt_adapter_request=prompt_adapter_request,
pooling_task="classify",
)
@@ -1267,7 +1238,6 @@ class LLM:
truncate_prompt_tokens: Optional[int] = None,
use_tqdm: Union[bool, Callable[..., tqdm]] = True,
lora_request: Optional[Union[list[LoRARequest], LoRARequest]] = None,
prompt_adapter_request: Optional[PromptAdapterRequest] = None,
) -> list[ScoringRequestOutput]:
encoded_output: list[PoolingRequestOutput] = self.encode(
@@ -1275,7 +1245,6 @@ class LLM:
truncate_prompt_tokens=truncate_prompt_tokens,
use_tqdm=use_tqdm,
lora_request=lora_request,
prompt_adapter_request=prompt_adapter_request,
pooling_task="embed",
)
@@ -1303,7 +1272,6 @@ class LLM:
truncate_prompt_tokens: Optional[int] = None,
use_tqdm: Union[bool, Callable[..., tqdm]] = True,
lora_request: Optional[Union[list[LoRARequest], LoRARequest]] = None,
prompt_adapter_request: Optional[PromptAdapterRequest] = None,
) -> list[ScoringRequestOutput]:
if isinstance(tokenizer, MistralTokenizer):
@@ -1361,7 +1329,6 @@ class LLM:
params=pooling_params,
use_tqdm=use_tqdm,
lora_request=lora_request,
prompt_adapter_request=prompt_adapter_request,
)
outputs = self._run_engine(use_tqdm=use_tqdm)
@@ -1381,7 +1348,6 @@ class LLM:
truncate_prompt_tokens: Optional[int] = None,
use_tqdm: Union[bool, Callable[..., tqdm]] = True,
lora_request: Optional[Union[list[LoRARequest], LoRARequest]] = None,
prompt_adapter_request: Optional[PromptAdapterRequest] = None,
) -> list[ScoringRequestOutput]:
"""Generate similarity scores for all pairs `<text,text_pair>` or
`<multi-modal data, multi-modal data pair>`.
@@ -1412,8 +1378,6 @@ class LLM:
it is used to create the progress bar.
If `False`, no progress bar is created.
lora_request: LoRA request to use for generation, if any.
prompt_adapter_request: Prompt Adapter request to use for
generation, if any.
Returns:
A list of `ScoringRequestOutput` objects containing the
@@ -1504,8 +1468,7 @@ class LLM:
data_2, # type: ignore[arg-type]
truncate_prompt_tokens,
use_tqdm,
lora_request,
prompt_adapter_request)
lora_request)
else:
return self._embedding_score(
tokenizer,
@@ -1513,8 +1476,7 @@ class LLM:
data_2, # type: ignore[arg-type]
truncate_prompt_tokens,
use_tqdm,
lora_request,
prompt_adapter_request)
lora_request)
def start_profile(self) -> None:
self.llm_engine.start_profile()
@@ -1625,7 +1587,6 @@ class LLM:
*,
use_tqdm: Union[bool, Callable[..., tqdm]] = True,
lora_request: Optional[Union[Sequence[LoRARequest], LoRARequest]],
prompt_adapter_request: Optional[PromptAdapterRequest],
tokenization_kwargs: Optional[dict[str, Any]] = None,
guided_options: Optional[GuidedDecodingRequest] = None,
priority: Optional[list[int]] = None,
@@ -1671,7 +1632,6 @@ class LLM:
tokenization_kwargs=tokenization_kwargs,
lora_request=lora_request[i] if isinstance(
lora_request, Sequence) else lora_request,
prompt_adapter_request=prompt_adapter_request,
priority=priority[i] if priority else 0,
)
@@ -1681,7 +1641,6 @@ class LLM:
params: Union[SamplingParams, PoolingParams],
tokenization_kwargs: Optional[dict[str, Any]] = None,
lora_request: Optional[LoRARequest] = None,
prompt_adapter_request: Optional[PromptAdapterRequest] = None,
priority: int = 0,
) -> None:
request_id = str(next(self.request_counter))
@@ -1691,7 +1650,6 @@ class LLM:
params,
lora_request=lora_request,
tokenization_kwargs=tokenization_kwargs,
prompt_adapter_request=prompt_adapter_request,
priority=priority,
)