Fix test_lora_qwen3 nightly failure: replace adapter with added_tokens

[alisonshao]

Motivation

Fix nightly 1-GPU test failure for test_lora_qwen3.py (exit code -9).

Failure: https://github.com/sgl-project/sglang/actions/runs/22026980777/job/63645203194

Modifications

Replace LoRA test adapter y9760210/Qwen3-4B-lora_model with TanXS/Qwen3-4B-LoRA-ZH-WebNovelty-v0.0 in LORA_MODELS_QWEN3.

Root cause: PR #18046 added validation in lora_manager.py that rejects LoRA adapters with added_tokens.json. The test adapter y9760210/Qwen3-4B-lora_model contains added_tokens.json, causing the server to crash during adapter loading.

Fix: Replace with TanXS/Qwen3-4B-LoRA-ZH-WebNovelty-v0.0, a valid Qwen3-4B LoRA adapter (rank 16, targets all attention + MLP modules) that does not have added_tokens.json.

[gemini-code-assist]

Summary of Changes

Hello @alisonshao, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request addresses a recurring nightly test failure in test_lora_qwen3.py. The issue stemmed from an incompatible LoRA adapter being used in the test suite, which caused server crashes due to recent validation changes. By substituting the problematic adapter with a valid one that adheres to the updated rules, the PR ensures the stability and reliability of the test environment, preventing future failures related to adapter loading.

Highlights

• Test Fix: Replaced an invalid LoRA adapter in the LORA_MODELS_QWEN3 configuration to resolve a nightly test failure in test_lora_qwen3.py.
• Root Cause: The previous adapter (y9760210/Qwen3-4B-lora_model) contained added_tokens.json, which is now rejected by validation added in lora_manager.py, causing server crashes during adapter loading.
• Solution: Swapped the problematic adapter with TanXS/Qwen3-4B-LoRA-ZH-WebNovelty-v0.0, a valid Qwen3-4B LoRA adapter that does not include added_tokens.json.

🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console.

Changelog

• python/sglang/test/lora_utils.py
  • Updated the LORA_MODELS_QWEN3 list to use a different Qwen3-4B LoRA adapter.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[gemini-code-assist]

Code Review

This pull request addresses a nightly test failure in test_lora_qwen3.py by replacing a LoRA test adapter that is no longer valid. The previous adapter included an added_tokens.json file, which is now rejected by the lora_manager. The fix involves substituting it with a valid adapter that does not contain this file. The change is correct and directly resolves the test failure. The implementation looks good.

[alisonshao]

testing at: https://github.com/sgl-project/sglang/actions/runs/22051897676/job/63711607206

[yushengsu-thu]

no problem.