Skip to content

Fix inference unit test #4589

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
maanug-nv merged 2 commits into from
May 4, 2026
+1 −4
Merged

Fix inference unit test #4589

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
1ff6a99
initialize model parallel
maanug-nv May 2, 2026
a9f7efe
Merge branch 'main' into fix-static-infer-test
ko3n1g May 2, 2026
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
5 changes: 1 addition & 4 deletions tests/unit_tests/test_inference.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -28,6 +28,7 @@ def gpt2_tiktoken_tokenizer():

@pytest.fixture(scope="module")
def static_inference_engine(gpt2_tiktoken_tokenizer):
Utils.initialize_model_parallel()
engine_wrapper = StaticInferenceEngineTestHarness()
engine_wrapper.setup_engine(vocab_size=gpt2_tiktoken_tokenizer.vocab_size, legacy=True)

Expand Down Expand Up @@ -63,8 +64,6 @@ def client(app):
'megatron.core.inference.text_generation_server.text_generation_server.send_do_generate'
)
def test_generations_endpoint(mock_send_do_generate, client, gpt2_tiktoken_tokenizer):
Utils.initialize_distributed()

prompts = ["twinkle twinkle little star, how I wonder what you are"]
request_data = {"prompts": prompts, "tokens_to_generate": 10, "logprobs": True}

Expand All @@ -90,8 +89,6 @@ def test_generations_endpoint(mock_send_do_generate, client, gpt2_tiktoken_token
"megatron.core.inference.text_generation_server.endpoints.completions.send_do_generate"
)
def test_completions_endpoint(mock_send_do_generate, client, gpt2_tiktoken_tokenizer):
Utils.initialize_distributed()

twinkle = ("twinkle twinkle little star,", " how I wonder what you are")
request_data = {"prompt": twinkle[0] + twinkle[1], "max_tokens": 0, "logprobs": 5, "echo": True}

Expand Down
Loading