Skip to content
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

Fix tokenizer for vllm worker #2984

Merged
merged 4 commits into from
Jan 31, 2024
Merged

Fix tokenizer for vllm worker #2984

Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
bcd88dd
fix tokenizer for vllm worker
Michaelvll Jan 31, 2024
9f2723b
backward compat
Michaelvll Jan 31, 2024
25c6578
fix comment
Michaelvll Jan 31, 2024
58440d8
format
Michaelvll Jan 31, 2024
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: 4 additions & 0 deletions fastchat/serve/vllm_worker.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -55,6 +55,10 @@ def __init__(
f"Loading the model {self.model_names} on worker {worker_id}, worker type: vLLM worker..."
)
self.tokenizer = llm_engine.engine.tokenizer
# This is to support vllm >= 0.2.7 where TokenizerGroup was introduced
# and llm_engine.engine.tokenizer was no longer a raw tokenizer
if hasattr(self.tokenizer, "tokenizer"):
self.tokenizer = llm_engine.engine.tokenizer.tokenizer
self.context_len = get_context_length(llm_engine.engine.model_config.hf_config)

if not no_register:
Expand Down
Loading