Skip to content

[Model] Support bitsandbytes for MiniCPMV #9891

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Nov 1, 2024
Merged

[Model] Support bitsandbytes for MiniCPMV #9891

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
43 changes: 43 additions & 0 deletions vllm/model_executor/models/minicpmv.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -810,6 +810,28 @@ class MiniCPMV2_5(MiniCPMVBaseModel, SupportsLoRA):
# resampler
"kv_proj",
]

# BitandBytes specific attributes
default_bitsandbytes_target_modules = [
".gate_proj.",
".down_proj.",
".up_proj.",
".q_proj.",
".k_proj.",
".v_proj.",
".o_proj.",
]
# in TP, these weights are partitioned along the column dimension (dim=-1)
column_parallel_weights_modules = [".down_proj.", ".o_proj."]
bitsandbytes_stacked_params_mapping = {
# shard_name, weight_name, index
"q_proj": ("qkv_proj", 0),
"k_proj": ("qkv_proj", 1),
"v_proj": ("qkv_proj", 2),
"gate_proj": ("gate_up_proj", 0),
"up_proj": ("gate_up_proj", 1),
}

embedding_modules = {}
embedding_padding_modules = []

Expand Down Expand Up @@ -931,6 +953,27 @@ class MiniCPMV2_6(MiniCPMVBaseModel, SupportsLoRA):
"kv_proj",
]

# BitandBytes specific attributes
default_bitsandbytes_target_modules = [
".gate_proj.",
".down_proj.",
".up_proj.",
".q_proj.",
".k_proj.",
".v_proj.",
".o_proj.",
]
# in TP, these weights are partitioned along the column dimension (dim=-1)
column_parallel_weights_modules = [".down_proj.", ".o_proj."]
bitsandbytes_stacked_params_mapping = {
# shard_name, weight_name, index
"q_proj": ("qkv_proj", 0),
"k_proj": ("qkv_proj", 1),
"v_proj": ("qkv_proj", 2),
"gate_proj": ("gate_up_proj", 0),
"up_proj": ("gate_up_proj", 1),
}

embedding_modules = {}
embedding_padding_modules = []

Expand Down