Skip to content

[CI] Fix flaky threshold #8370

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
merrymercy merged 1 commit into from
Jul 25, 2025
Merged

[CI] Fix flaky threshold #8370

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: 2 additions & 2 deletions test/srt/test_vllm_dependency.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17,8 +17,8 @@
)

MODEL_SCORE_THRESHOLDS = {
"hugging-quants/Meta-Llama-3.1-8B-Instruct-AWQ-INT4": 0.83,
"hugging-quants/Meta-Llama-3.1-8B-Instruct-GPTQ-INT4": 0.83,
"hugging-quants/Meta-Llama-3.1-8B-Instruct-AWQ-INT4": 0.825,
"hugging-quants/Meta-Llama-3.1-8B-Instruct-GPTQ-INT4": 0.825,
Comment on lines +20 to +21
Copy link
Copy Markdown
Contributor

@gemini-code-assist gemini-code-assist bot Jul 25, 2025

Choose a reason for hiding this comment

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

medium

Lowering test thresholds can sometimes mask underlying issues. Could you please add more context to the pull request description about why these tests are flaky? For example, what was the range of scores you observed in failing test runs?

To improve maintainability, it would be very helpful to add a comment explaining why this threshold was chosen. This will help future developers understand the context behind this value and prevent them from changing it arbitrarily. Please also add the PR number to the comment once it's available.

Suggested change
"hugging-quants/Meta-Llama-3.1-8B-Instruct-AWQ-INT4": 0.825,
"hugging-quants/Meta-Llama-3.1-8B-Instruct-GPTQ-INT4": 0.825,
"hugging-quants/Meta-Llama-3.1-8B-Instruct-AWQ-INT4": 0.825, # Lowered from 0.83 to address flakiness. See PR #...
"hugging-quants/Meta-Llama-3.1-8B-Instruct-GPTQ-INT4": 0.825, # Lowered from 0.83 to address flakiness. See PR #...

"hugging-quants/Mixtral-8x7B-Instruct-v0.1-AWQ-INT4": 0.62,
}

Expand Down
Loading