Skip to content

[Models] Use in-place adds in Idefics2Vision #23932

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
Aug 29, 2025
Merged

[Models] Use in-place adds in Idefics2Vision #23932

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
6 changes: 3 additions & 3 deletions vllm/model_executor/models/idefics2_vision_model.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -108,7 +108,7 @@ def forward(self,
bucket_coords_w).flatten()
position_ids[batch_idx][p_attn_mask.view(-1).cpu()] = pos_ids
position_ids = position_ids.to(self.position_embedding.weight.device)
embeddings = embeddings + self.position_embedding(position_ids)
embeddings += self.position_embedding(position_ids)
return embeddings


Expand Down Expand Up @@ -262,11 +262,11 @@ def forward(
residual = hidden_states
hidden_states = self.layer_norm1(hidden_states)
hidden_states = self.self_attn(hidden_states)
hidden_states = residual + hidden_states
hidden_states += residual
residual = hidden_states
hidden_states = self.layer_norm2(hidden_states)
hidden_states = self.mlp(hidden_states)
hidden_states = residual + hidden_states
hidden_states += residual
return hidden_states


Expand Down