75 lines
2.2 KiB
YAML
75 lines
2.2 KiB
YAML
name: Bug Report
|
|
description: Report a bug or unexpected behavior
|
|
labels: ["bug"]
|
|
title: "bug: <short summary>"
|
|
body:
|
|
- type: markdown
|
|
attributes:
|
|
value: |
|
|
🐞 **Thanks for reporting a bug!**
|
|
|
|
To help us investigate and fix the issue efficiently, please provide as much
|
|
relevant information as possible. Clear and reproducible reports are highly appreciated.
|
|
|
|
- type: textarea
|
|
attributes:
|
|
label: Bug Description
|
|
description: |
|
|
Clearly and concisely describe the bug.
|
|
What happened? What is broken or behaving incorrectly?
|
|
placeholder: |
|
|
Example:
|
|
- vLLM crashes when loading model XXX
|
|
- Unexpected latency spike during decode stage
|
|
validations:
|
|
required: true
|
|
|
|
- type: textarea
|
|
attributes:
|
|
label: Steps to Reproduce
|
|
description: |
|
|
Provide the exact steps to reproduce the issue.
|
|
Please include commands, configuration, and minimal repro if possible.
|
|
placeholder: |
|
|
Example:
|
|
1. Start vLLM with config XXX
|
|
2. Send request YYY
|
|
3. Observe error or incorrect behavior
|
|
validations:
|
|
required: true
|
|
|
|
- type: textarea
|
|
attributes:
|
|
label: Expected Behavior
|
|
description: |
|
|
Describe what you expected to happen instead.
|
|
This helps clarify whether the behavior is incorrect or just unexpected.
|
|
placeholder: |
|
|
Example:
|
|
- Model should load successfully
|
|
- Latency should remain stable under N requests
|
|
validations:
|
|
required: false
|
|
|
|
- type: textarea
|
|
attributes:
|
|
label: Additional Context
|
|
description: |
|
|
Add any additional information that may help diagnose the issue.
|
|
This can include logs, stack traces, environment details, or related issues.
|
|
placeholder: |
|
|
- Logs / stack traces
|
|
- OS, CUDA, driver, hardware info
|
|
- vLLM / Kunlun version
|
|
- Related issues or PRs
|
|
validations:
|
|
required: false
|
|
|
|
- type: markdown
|
|
attributes:
|
|
value: |
|
|
👍 **Does this bug affect you as well?**
|
|
|
|
Please consider giving it a 👍.
|
|
We often prioritize issues that impact a larger portion of the community.
|