Skip to content

[Test] Skip STANDALONE speculative decoding tests for different hidden sizes #14733

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
hnyls2002 merged 1 commit into from
Dec 9, 2025
Merged

[Test] Skip STANDALONE speculative decoding tests for different hidden sizes #14733

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
8 changes: 8 additions & 0 deletions test/srt/models/test_nvidia_nemotron_nano_v2.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -24,6 +24,10 @@ class TestNvidiaNemotronNanoV2NVFP4(GSM8KMixin, CustomTestCase):
other_args = ["--max-mamba-cache-size", "256"]


@unittest.skip(
"STANDALONE speculative decoding does not yet support target and draft models "
"with different hidden sizes (Nemotron-9B: 4480, Llama-3.2-1B: 2048)"
)
Comment on lines +27 to +30
Copy link
Copy Markdown
Contributor

@gemini-code-assist gemini-code-assist bot Dec 9, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

medium

The skip reason message is duplicated for the TestNvidiaNemotronNanoV2SpeculativeDecodingBF16Cache test class on line 56. To improve maintainability and follow the DRY (Don't Repeat Yourself) principle, it would be better to define this string as a module-level constant and reuse it in both @unittest.skip decorators.

For example, you could define the following constant at the top of the file:

_STANDALONE_SPEC_DECODING_SKIP_REASON = (
    "STANDALONE speculative decoding does not yet support target and draft models "
    "with different hidden sizes (Nemotron-9B: 4480, Llama-3.2-1B: 2048)"
)

Then, you can use it like this:

@unittest.skip(_STANDALONE_SPEC_DECODING_SKIP_REASON)
class TestNvidiaNemotronNanoV2SpeculativeDecoding(GSM8KMixin, CustomTestCase):
    # ...

@unittest.skip(_STANDALONE_SPEC_DECODING_SKIP_REASON)
class TestNvidiaNemotronNanoV2SpeculativeDecodingBF16Cache(GSM8KMixin, CustomTestCase):
    # ...

class TestNvidiaNemotronNanoV2SpeculativeDecoding(GSM8KMixin, CustomTestCase):
accuracy = 0.87
model = "nvidia/NVIDIA-Nemotron-Nano-9B-v2"
Expand All @@ -49,6 +53,10 @@ class TestNvidiaNemotronNanoV2SpeculativeDecoding(GSM8KMixin, CustomTestCase):
]


@unittest.skip(
"STANDALONE speculative decoding does not yet support target and draft models "
"with different hidden sizes (Nemotron-9B: 4480, Llama-3.2-1B: 2048)"
)
class TestNvidiaNemotronNanoV2SpeculativeDecodingBF16Cache(GSM8KMixin, CustomTestCase):
accuracy = 0.87
model = "nvidia/NVIDIA-Nemotron-Nano-9B-v2"
Expand Down
Loading