vLLM is a library for LLM inference and serving. From 0.12.0 to before 0.24.0, sending a pure prompt embeds payload in a /v1/completions request with a model using M-RoPE causes EngineCore to fail an assertion and fatally crash, shutting down the entire server application. Any remote user who is authorized to make a /v1/completions request can make such a request and induce a crash. This issue is fixed in version 0.24.0.
CVE-2026-55514
This medium-severity CVE scores 6.5 under NVD CVSS v3. EPSS exploit probability: 0.4%, top 71% of all CVEs by exploit prediction. GitHub Security Advisory data not yet ingested — confidence will rise once GHSA publishes (typical lag: hours to days for open-source ecosystem CVEs; never for infrastructure-only CVEs).
- Lower severity and no public exploit yet
No vendor fix yet — apply a workaround or compensating control (WAF / firewall / segmentation) and watch for a patch.
- CVSS v3
- 6.5
- EG Score
- 6.5(medium)
- EPSS
- 29.1%
- KEV
- Not listed
Published
July 6, 2026
Last Modified
July 7, 2026
Advisory Details (4)
Auto-updated Jul 7, 2026DoS caused by sending `/v1/completions` with prompt embeds payload with models that use M-RoPE · Advisory · vllm-project/vllm · GitHub
https://github.com/vllm-project/vllm/security/advisories/GHSA-33cg-gxv8-3p8gv0.24.0
Patch available: vllm-project/vllm v0.24.0
https://github.com/vllm-project/vllm/releases/tag/v0.24.0[Security] Fix DoS via prompt_embeds on M-RoPE models
Fix merged in vllm-project/vllm PR #45252 on 2026-06-13 — awaiting tagged release
https://github.com/vllm-project/vllm/pull/45252commit 470229c37efa (vllm-project/vllm)
Fix landed in vllm-project/vllm commit 470229c37efa — awaiting tagged release
https://github.com/vllm-project/vllm/commit/470229c37efaf69c86e8bc97482b0b1ff7551c65Weakness Classification(1)
MITRE Common Weakness Enumeration — the root-cause categories this CVE belongs to.
Data Freshness Timeline
(refreshed 18× in last 7d / 18× in last 30d)
Each row is a source pipeline that fetched or updated this CVE on that date, with what changed. For example, "NVD update" means NVD published or revised its analysis for this CVE; "MITRE cvelistV5" means we ingested or refreshed it from the CNA feed. Most recent first.
- 2026-07-12 05:46 UTCEPSS rescore
- 2026-07-11 23:52 UTCEG score recompute
- 2026-07-11 08:27 UTCEPSS rescore
- 2026-07-11 08:27 UTCEPSS rescore
- 2026-07-10 23:20 UTCEG score recompute
- 2026-07-09 22:48 UTCEG score recompute
- 2026-07-09 19:10 UTCEPSS rescore
- 2026-07-09 19:10 UTCEPSS rescore
- 2026-07-08 22:16 UTCEG score recompute
- 2026-07-08 15:16 UTCEPSS rescore
- 2026-07-08 15:16 UTCEPSS rescore
- 2026-07-07 21:43 UTCEG score recompute▼ 0.60
- 2026-07-07 20:14 UTCNVD updateCVSS v3 → 6.5 · severity → MEDIUM
- 2026-07-07 13:46 UTCEPSS rescore
- 2026-07-07 13:46 UTCEPSS rescore
- 2026-07-07 09:06 UTCEG score recompute
- 2026-07-06 20:30 UTCEG score recompute
- 2026-07-06 20:29 UTCMITRE cvelistV5first tracked
Related CVEs(same CWE)
Frequently asked(5)
What is CVE-2026-55514?
When was CVE-2026-55514 disclosed?
Is CVE-2026-55514 actively exploited?
What is the CVSS score of CVE-2026-55514?
How do I remediate CVE-2026-55514?
Dependency Blast Radius
Explore the affected products and dependency analysis for CVE-2026-55514
Is Your Infrastructure Affected by CVE-2026-55514?
EchelonGraph automatically scans your cloud infrastructure and maps CVE exposure using blast radius analysis.