Issue Station
Sign In
Tools hereToolbox

vllm - 💡(How to fix) Fix [Bug]: Runtime LoRA unload does not remove adapter from engine [1 pull requests]

Official PRs (…)
ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

Helpful · Quick feedback

Loading…

When using /v1/load_lora_adapter, the adapter is registered in the OpenAI serving layer and also preloaded into the engine via add_lora.

However, /v1/unload_lora_adapter currently only removes the adapter from the OpenAI serving-side registry. It does not call the engine-side remove_lora, so the engine-side LoRA manager may still retain the adapter after the API reports that it has been unloaded.

Root Cause

When using /v1/load_lora_adapter, the adapter is registered in the OpenAI serving layer and also preloaded into the engine via add_lora.

However, /v1/unload_lora_adapter currently only removes the adapter from the OpenAI serving-side registry. It does not call the engine-side remove_lora, so the engine-side LoRA manager may still retain the adapter after the API reports that it has been unloaded.

Fix Action

Fixed

Code Example

Your output of `python collect_env.py` here
RAW_BUFFERClick to expand / collapse

Your current environment

<details> <summary>The output of <code>python collect_env.py</code></summary> 
Your output of `python collect_env.py` here
</details>

🐛 Describe the bug

Bug

Runtime LoRA unload does not fully remove the adapter from the engine.

Description

When using /v1/load_lora_adapter, the adapter is registered in the OpenAI serving layer and also preloaded into the engine via add_lora.

However, /v1/unload_lora_adapter currently only removes the adapter from the OpenAI serving-side registry. It does not call the engine-side remove_lora, so the engine-side LoRA manager may still retain the adapter after the API reports that it has been unloaded.

Expected behavior

Calling /v1/unload_lora_adapter should remove the LoRA adapter from both:

  • the OpenAI serving-side registry
  • the engine-side LoRA manager

Related discussion

Originally noticed while looking into #42207.

Before submitting a new issue...

  • Make sure you already searched for relevant issues, and asked the chatbot living at the bottom right corner of the documentation page, which can answer lots of frequently asked questions.

Vote matrix · Quick signals

Works
Did the solution work? Tap to confirm.
Easy Fix
Was it a quick fix?
Time Saver
Did it save you time?
Blocking
Was it severely blocking?
Common Issue
Are others likely hitting this too?
Flaky / Intermittent
Is it intermittent?
Verified / Reproducible
Can you reproduce it reliably?
Loading…

FAQ

Expected behavior

Calling /v1/unload_lora_adapter should remove the LoRA adapter from both:

  • the OpenAI serving-side registry
  • the engine-side LoRA manager
#api#latency issue#model loading#dependency error#configuration error

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Back to top recommendations

TRENDING

vllm - 💡(How to fix) Fix [Bug]: Runtime LoRA unload does not remove adapter from engine [1 pull requests]