CVE-2026-115917

vLLM is an inference and serving engine for large language models. Prior to 0.32.0, the OpenAI-compatible server configures Starlette CORSMiddleware with allowed_origins set to a wildcard by default. A malicious web page can use a victim's browser to reach an unauthenticated vLLM server on localhost, an internal network, or a VPN, call endpoints including GET /v1/models and POST /v1/chat/completions, consume GPU resources, and read responses. Cookie-based cross-site request forgery does not apply because credentials are disabled by default, and deployments requiring bearer authentication are exposed only when the caller already has a usable token. This issue is fixed in version 0.32.0.
Configurations

Configuration 1 (hide)

cpe:2.3:a:vllm:vllm:*:*:*:*:*:*:*:*

History

28 Oct 2026, 15:35

Type Values Removed Values Added
References () https://github.com/vllm-project/vllm/commit/fcae5c98870ff2c20659731dce7ba3a3ff0d48aa - () https://github.com/vllm-project/vllm/commit/fcae5c98870ff2c20659731dce7ba3a3ff0d48aa - Patch
References () https://github.com/vllm-project/vllm/pull/61009 - () https://github.com/vllm-project/vllm/pull/61009 - Patch
References () https://github.com/vllm-project/vllm/releases/tag/v0.32.0 - () https://github.com/vllm-project/vllm/releases/tag/v0.32.0 - Release Notes
References () https://github.com/vllm-project/vllm/security/advisories/GHSA-vvpg-rrww-wpmh - () https://github.com/vllm-project/vllm/security/advisories/GHSA-vvpg-rrww-wpmh - Patch, Vendor Advisory
First Time Vllm vllm
Vllm
CPE cpe:2.3:a:vllm:vllm:*:*:*:*:*:*:*:*

26 Oct 2026, 18:17

Type Values Removed Values Added
New CVE

Information

Published : 2026-10-26 18:17

Updated : 2026-10-31 02:17


NVD link : CVE-2026-115917

Mitre link : CVE-2026-115917

CVE.ORG link : CVE-2026-115917


JSON object : View

Products Affected

vllm

  • vllm
CWE
CWE-942

Permissive Cross-domain Security Policy with Untrusted Domains