xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-07-17 02:37:09 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Quant] Bamba SupportsQuant (#14698)

Browse Source 
Signed-off-by: Kyle Sayers <kylesayrs@gmail.com>
This commit is contained in:
Kyle Sayers 2025-03-13 00:57:05 -04:00 committed by GitHub
parent 36d1ccb286
commit 5d043c1685
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
vllm/model_executor/models/bamba.py
View File

@ -33,7 +33,7 @@ from vllm.sequence import IntermediateTensors
from vllm.utils import LayerBlockType from vllm.utils import LayerBlockType
from .interfaces import (HasInnerState, IsHybrid, SupportsLoRA, SupportsPP, from .interfaces import (HasInnerState, IsHybrid, SupportsLoRA, SupportsPP,
SupportsV0Only) SupportsQuant, SupportsV0Only)
from .utils import (is_pp_missing_parameter, from .utils import (is_pp_missing_parameter,
make_empty_intermediate_tensors_factory, make_layers, make_empty_intermediate_tensors_factory, make_layers,
maybe_prefix) maybe_prefix)
@ -367,7 +367,7 @@ class BambaModel(nn.Module):
class BambaForCausalLM(nn.Module, HasInnerState, SupportsLoRA, SupportsPP, class BambaForCausalLM(nn.Module, HasInnerState, SupportsLoRA, SupportsPP,
IsHybrid, SupportsV0Only): IsHybrid, SupportsV0Only, SupportsQuant):
packed_modules_mapping = { packed_modules_mapping = {
"qkv_proj": [ "qkv_proj": [
"q_proj", "q_proj",