What is GHSA-pjrq-r86m-9r53?

GHSA-pjrq-r86m-9r53 is a security advisory published by GitHub Security Advisories with High severity. vLLM from 0.28.0 before 0.30.0 contains a reachable assertion vulnerability in GPUModelRunner...

Which CVE IDs does GHSA-pjrq-r86m-9r53 cover?

GHSA-pjrq-r86m-9r53 covers the following CVE IDs: CVE-2026-126398. Each CVE has its own detail page on EchelonGraph Pulse with the synthesized EG score.

What is the CVSS v3 score of GHSA-pjrq-r86m-9r53?

GHSA-pjrq-r86m-9r53 has a CVSS v3 base score of 5.9, published by GitHub Security Advisories.

GHSA-pjrq-r86m-9r53HighCVSS 5.9

vLLM from 0.28.0 before 0.30.0 contains a reachable assertion vulnerability in GPUModelRunner...

Vendor

GitHub Security Advisories

Published

November 13, 2026

Last Modified

November 13, 2026

🔗 CVE IDs covered (1)

CVE-2026-126398 →

📋 Description

vLLM from 0.28.0 before 0.30.0 contains a reachable assertion vulnerability in GPUModelRunner._init_xdrope_positions that allows remote attackers to crash the engine by submitting embeddings-only requests. Attackers can send a /v1/completions request containing only prompt_embeds to XD-RoPE models with --enable-prompt-embeds, terminating the EngineCore and API server for all clients.

🔗 References (4)

https://github.com/vllm-project/vllm/security/advisories/GHSA-5q78-j82c-2vr2
https://nvd.nist.gov/vuln/detail/CVE-2026-126398
https://www.vulncheck.com/advisories/vllm-0.28.0-before-0.30.0-denial-of-service-via-prompt-embeds-on-xd-rope-models
https://github.com/advisories/GHSA-pjrq-r86m-9r53