Skip to content

[CI Bugfix] Fix failing test_model_load_with_params tests due to tokenizer refactor #25086

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
Sep 17, 2025
Merged

[CI Bugfix] Fix failing test_model_load_with_params tests due to tokenizer refactor #25086

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
11 changes: 5 additions & 6 deletions tests/model_executor/test_model_load_with_params.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -47,8 +47,8 @@ def test_model_loading_with_params(vllm_runner, monkeypatch):
assert model_config.pooler_config.normalize

# asserts on the tokenizer loaded
assert model_tokenizer.tokenizer_id == "BAAI/bge-base-en-v1.5"
assert model_tokenizer.tokenizer.model_max_length == 512
assert model_config.tokenizer == "BAAI/bge-base-en-v1.5"
assert model_tokenizer.model_max_length == 512

def check_model(model):
assert isinstance(model, BertEmbeddingModel)
Expand Down Expand Up @@ -87,8 +87,8 @@ def test_roberta_model_loading_with_params(vllm_runner, monkeypatch):
assert model_config.pooler_config.normalize

# asserts on the tokenizer loaded
assert model_tokenizer.tokenizer_id == "intfloat/multilingual-e5-base"
assert model_tokenizer.tokenizer.model_max_length == 512
assert model_config.tokenizer == "intfloat/multilingual-e5-base"
assert model_tokenizer.model_max_length == 512

def check_model(model):
assert isinstance(model, RobertaEmbeddingModel)
Expand Down Expand Up @@ -116,8 +116,7 @@ def test_facebook_roberta_model_loading_with_params(vllm_runner, monkeypatch):
output = vllm_model.embed("Write a short story about a robot that"
" dreams for the first time.\n")

model_tokenizer = vllm_model.llm.llm_engine.tokenizer
assert model_tokenizer.tokenizer_id == model_name
assert vllm_model.llm.llm_engine.model_config.tokenizer == model_name

def check_model(model):
assert isinstance(model, RobertaEmbeddingModel)
Expand Down