CVE-2026-37237: Vllm vllm vulnerability
vLLM up to and including 0.17.0 allows remote attackers to cause a Denial of Service via memory exhaustion. The AsyncMediaIO.fetchaudio and AsyncMediaIO.fetchimage functions in multimodal/inputs.py fetch user-supplied media URLs using aiohttp and call r.read() without enforcing a maximum response size, allowing an attacker to exhaust server memory by providing a URL to an arbitrarily large file.
Affected Software
Event History
Frequently Asked Questions
Which deployments are exposed to this issue?
vLLM deployments up to and including version 0.17.0 are affected when they use the asynchronous multimodal media-fetching paths for user-supplied audio or image URLs.
What does an attacker need to do to trigger the denial of service?
An attacker needs to provide a media URL that resolves to an arbitrarily large file. The affected fetch functions read the full response without a maximum response-size limit, which can exhaust server memory.
What is the impact if exploitation succeeds?
The attacker can cause memory exhaustion on the vLLM server, resulting in a denial of service.