[NEW] [medium] vllm: Multiple vulnerabilities allow Denial of Service
An attacker can exploit multiple vulnerabilities in vllm to perform a Denial of Service attack.
CSIRTS triage
- What
- Multiple vulnerabilities allow an attacker to perform a Denial of Service attack.
- Who is affected
- Attackers exploiting vulnerabilities in vllm.
- Urgency
- Remediation is medium urgency due to the potential for Denial of Service attacks.
- Action
- Apply the latest patches for vllm.
AI-assisted analysis generated from the source advisory — verify against the original.
⚡ Watch vllm
Get an email when a new vllm advisory drops — max one per day, one-click unsubscribe.
Details
Original advisory: https://wid.cert-bund.de/portal/wid/securityadvisory?name=WID-SEC-2026-2176
Exploitation outlook
EPSS (FIRST.org) estimates each CVE’s probability of exploitation in the next 30 days — here is the CSIRTS.com read on those numbers.
- Low exploitation riskCVE-2026-542340.62% 30-day exploitation probability — currently an unlikely target, but scores change as exploit code circulates. Riskier than 47% of all EPSS-scored CVEs.
- Low exploitation riskCVE-2026-555140.66% 30-day exploitation probability — currently an unlikely target, but scores change as exploit code circulates. Riskier than 49% of all EPSS-scored CVEs.
- Low exploitation riskCVE-2026-555740.58% 30-day exploitation probability — currently an unlikely target, but scores change as exploit code circulates. Riskier than 45% of all EPSS-scored CVEs.
- Low exploitation riskCVE-2026-556460.52% 30-day exploitation probability — currently an unlikely target, but scores change as exploit code circulates. Riskier than 42% of all EPSS-scored CVEs.
Referenced CVEs
| CVE | CSIRTS overview | External |
|---|---|---|
| CVE-2026-54234 | coverage & exploitation status | NVD · CVE.org |
| CVE-2026-55514 | coverage & exploitation status | NVD · CVE.org |
| CVE-2026-55574 | coverage & exploitation status | NVD · CVE.org |
| CVE-2026-55646 | coverage & exploitation status | NVD · CVE.org |
Same CVEs, other sources
How other CERTs, PSIRTs and databases cover the vulnerabilities in this advisory.
- highGHSA-33cg-gxv8-3p8g: vLLM denial of service via prompt embeds on M-RoPE modelsghsa
- mediumGHSA-v82g-2437-67m2: vLLM: Speech-to-text upload size limit is enforced after full UploadFile readghsa
- highGHSA-rwxx-mrjm-wc2m: vLLM: ReDoS via structured_outputs.regex compiled without timeout in xgrammar and outline…ghsa
- highGHSA-8wr5-jm2h-8r4f: vLLM has Remote DoS via Invalid Recovered Token Reinjectionghsa
- highCVE-2026-55574: vLLM is a high-throughput and memory-efficient inference and serving engine for LLMs. Prior to…nvd
- mediumCVE-2026-55514: vLLM is a library for LLM inference and serving. From 0.12.0 to before 0.24.0, sending a pure …nvd
- highCVE-2026-54234: vLLM is a high-throughput and memory-efficient inference and serving engine for LLMs. Prior to…nvd
- mediumCVE-2026-55646: vLLM is an inference and serving engine for large language models. From 0.22.0 to 0.23.0, the …nvd
More from CERT-Bund (BSI) Security Advisories
- high[NEW] [high] Linux Kernel: Multiple vulnerabilities2026-08-25
- medium[NEW] [medium] libTIFF: Multiple Vulnerabilities2026-08-25
- high[NEW] [high] Contao: Multiple Vulnerabilities2026-08-25
- medium[NEW] [medium] Django: Multiple Vulnerabilities2026-08-25
- high[NEW] [high] Red Hat Enterprise Linux (Apicurio Registry): Multiple Vulnerabilities2026-08-25