Skip to content

fix: pass rope_scaling=None for Qwen3 to avoid unhashable dict error #198

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

Open
CruxZhou wants to merge 1 commit into
base: main
Choose a base branch
from
Open

fix: pass rope_scaling=None for Qwen3 to avoid unhashable dict error #198

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
4 changes: 3 additions & 1 deletion nanovllm/models/qwen3.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -132,7 +132,9 @@ def __init__(
qkv_bias=getattr(config, 'attention_bias', True),
head_dim=getattr(config, 'head_dim', None),
rope_theta=getattr(config, "rope_theta", 1000000),
rope_scaling=getattr(config, "rope_scaling", None),
rope_scaling=None,
# RoPE scaling is not implemented yet in nano-vllm rotary_embedding.get_rope,
# so pass None explicitly to avoid unhashable dict errors with newer transformers.
)
self.mlp = Qwen3MLP(
hidden_size=config.hidden_size,
Expand Down