[V1] Get supported tasks from model runner instead of model config (#21585)
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
This commit is contained in:
@@ -26,8 +26,8 @@ from vllm.model_executor.layers.quantization import QuantizationConfig
|
||||
from vllm.model_executor.layers.vocab_parallel_embedding import (
|
||||
VocabParallelEmbedding)
|
||||
from vllm.model_executor.pooling_metadata import PoolingMetadata
|
||||
from vllm.pooling_params import PoolingTask
|
||||
from vllm.sequence import IntermediateTensors
|
||||
from vllm.tasks import PoolingTask
|
||||
|
||||
from .interfaces import SupportsCrossEncoding, SupportsQuant, SupportsV0Only
|
||||
from .utils import AutoWeightsLoader, WeightsMapper, maybe_prefix
|
||||
|
||||
@@ -16,8 +16,8 @@ from vllm.model_executor.layers.pooler import (DispatchPooler, Pooler,
|
||||
get_prompt_token_ids)
|
||||
from vllm.model_executor.models.llama import LlamaForCausalLM
|
||||
from vllm.model_executor.pooling_metadata import PoolingMetadata
|
||||
from vllm.pooling_params import PoolingTask
|
||||
from vllm.sequence import PoolerOutput
|
||||
from vllm.tasks import PoolingTask
|
||||
from vllm.transformers_utils.tokenizer import cached_tokenizer_from_config
|
||||
|
||||
from .interfaces import SupportsV0Only
|
||||
|
||||
@@ -23,8 +23,8 @@ from vllm.model_executor.layers.vocab_parallel_embedding import (
|
||||
VocabParallelEmbedding)
|
||||
from vllm.model_executor.model_loader.weight_utils import default_weight_loader
|
||||
from vllm.model_executor.pooling_metadata import PoolingMetadata
|
||||
from vllm.pooling_params import PoolingTask
|
||||
from vllm.sequence import IntermediateTensors
|
||||
from vllm.tasks import PoolingTask
|
||||
|
||||
from .interfaces import SupportsCrossEncoding, SupportsV0Only
|
||||
from .utils import WeightsMapper, maybe_prefix
|
||||
|
||||
Reference in New Issue
Block a user