[SupportsQuant] Bert, Blip, Blip2, Bloom (#15573)

Signed-off-by: Kyle Sayers <kylesayrs@gmail.com>
This commit is contained in:
Kyle Sayers
2025-04-03 11:23:19 -04:00
committed by GitHub
parent 84884cd9ac
commit 421c462948
4 changed files with 16 additions and 9 deletions

View File

@@ -42,7 +42,7 @@ from vllm.model_executor.model_loader.weight_utils import default_weight_loader
from vllm.model_executor.sampling_metadata import SamplingMetadata
from vllm.sequence import IntermediateTensors
from .interfaces import SupportsPP, SupportsV0Only
from .interfaces import SupportsPP, SupportsQuant, SupportsV0Only
from .utils import (is_pp_missing_parameter,
make_empty_intermediate_tensors_factory, make_layers,
maybe_prefix)
@@ -279,7 +279,7 @@ class BloomModel(nn.Module):
return hidden_states
class BloomForCausalLM(nn.Module, SupportsPP, SupportsV0Only):
class BloomForCausalLM(nn.Module, SupportsPP, SupportsV0Only, SupportsQuant):
def __init__(self, *, vllm_config: VllmConfig, prefix: str = ""):
super().__init__()