Live › vulnerability › GHSA-3cgq-c484-rx5j
Go to dashboardvLLM, an LLM serving library, accepts per-request mm_processor_kwargs without validation, so a remote client can set huge image, patch, crop or frame counts. This exhausts memory and can crash the shared server for all users.
Summary written from the sources below. Check the originals before acting.
First seen by InfraStack 2026-10-23 20:59 UTC. All times on this page are UTC.
| Product | Ecosystem | Affected | Fixed in |
|---|---|---|---|
| vllm | pypi | >= 0.7.3, < 0.31.0 | – |
Run pip show vllm and check whether the version is >= 0.7.3 and < 0.31.0.
Upgrade vllm to 0.31.0 or later, per the duplicate advisory GHSA-gx7p-2j49-hfq4 which this entry points to.
Each row is the first time that source carried this signal. The analyst’s rows are its own findings.
This advisory has been withdrawn because it is a duplicate of GHSA-gx7p-2j49-hfq4. This link is maintained to preserve external references.
vLLM from 0.7.3 before 0.31.0 contains an uncontrolled resource consumption vulnerability that allows remote clients to control multimodal processor parameters by supplying per-request mm_processor_kwargs. Attackers can set large image size, patch count, crop count, or frame count values in Isaac, InternVL, H2OVL, and other processors to exhaust memory and terminate the shared server.
This page as .md · .json · in your agent: get_signal("GHSA-3cgq-c484-rx5j")
GitHub Advisory Database, CC-BY 4.0. Posts quoted belong to their authors.