LiveActive security incident?Get immediate response
CVE Record

CVE-2026-34756: vLLM Affected by Unauthenticated OOM Denial of Service via Unbounded `n` Parameter in OpenAI API Server

vLLM is an inference and serving engine for large language models (LLMs). From 0.1.0 to before 0.19.0, a Denial of Service vulnerability exists in the vLLM OpenAI-compatible API server. Due to the lack of an upper bound validation on the n parameter in the ChatCompletionRequest and CompletionRequest Pydantic models, an unauthenticated attacker can send a single HTTP request with an astronomically large n value. This completely blocks the Python asyncio event loop and causes immediate Out-Of-Memory crashes by allocating millions of request object copies in the heap before the request even reaches the scheduling queue. This vulnerability is fixed in 0.19.0.

MediumCVSS 6.5Not KEV-listedUpdated
Discuss exposureBack to CVEs
Glexia's TakeAutomated analysismoderate

Security readout for executives and security teams

Plain-English summary

A single malicious request to a vulnerable vLLM API can exhaust server memory and interrupt model-serving availability. The issue does not expose or alter data, but public-facing AI services could suffer immediate outages. vLLM versions 0.1.0 through 0.18.x are affected; version 0.19.0 fixes the flaw.

Executive priority

Treat this as a prompt availability remediation for exposed production AI services. Schedule rapid upgrades for internet-facing or business-critical endpoints, while using access restrictions as interim risk reduction. The medium severity does not reflect data theft, but one request may cause immediate service loss.

Technical view

The OpenAI-compatible server fails to impose an upper bound on the n field in chat-completion and completion request models. An extreme value causes massive request-object duplication before scheduling, blocking the Python asyncio event loop and exhausting heap memory. The reported CVSS 3.1 score is 6.5, with availability impact rated high.

Likely exposure

Exposure exists where vLLM 0.1.0 through 0.18.x runs its OpenAI-compatible API server and an untrusted party can submit completion requests. Internet-facing or broadly accessible internal endpoints present the greatest operational risk. Deployments on version 0.19.0 or later are not listed as affected.

Exploitation context

The source bundle describes a low-complexity, single-request denial of service. However, it provides no evidence of active exploitation, and the CVE is not identified as KEV. The advisory calls the attack unauthenticated, while the supplied CVSS vector states low privileges are required; that inconsistency should be considered when assessing reachability.

Researcher notes

CWE-1284 and CWE-770 align with improper input validation and uncontrolled resource consumption. Allocation reportedly occurs before requests reach the scheduling queue, so downstream capacity controls may not prevent failure. The bundle does not establish exploitation in the wild. Analysts should also reconcile the unauthenticated advisory language with CVSS PR:L.

Mitigation direction

  • Upgrade affected vLLM deployments to version 0.19.0 or later.
  • Prioritize externally reachable OpenAI-compatible API servers.
  • Restrict API access to trusted networks or authenticated clients until upgraded.
  • Apply vendor-provided Red Hat updates where those packaged products are deployed.
  • Monitor memory exhaustion, worker restarts, and sudden API unavailability.

Validation and detection

  • Inventory vLLM versions and identify deployments below 0.19.0.
  • Confirm whether chat-completion or completion endpoints accept untrusted requests.
  • Verify upgraded instances report vLLM 0.19.0 or later.
  • Safely confirm oversized n values are rejected in an isolated staging environment.
  • Review telemetry for memory spikes, event-loop stalls, and repeated service restarts.
Prepared
Confidence
high
Sources
6

Public sources used

Generated from the cited source records. This long-tail analysis has not been individually reviewed by a named human.

Potential ATT&CK relevance

Conservative CVE-to-ATT&CK context

These mappings and lookup hints may be relevant to the vulnerability behavior, CWE, affected product, or exposure path. Glexia-inferred context is not an official MITRE, ATT&CK, CWE, or CVE Program mapping.

ATT&CK lookup starting points

Use these exact CWE pages and searches to review the Glexia ATT&CK library from this CVE's weakness and description context.

cwe · low confidence lookup

CWE-1284: Exact CWE lookup

Use the exact CWE identifier as the starting point before reviewing related ATT&CK behavior. Open the exact CWE lookup page first, then review the ATT&CK searches from that MITRE weakness context. This is a Glexia lookup hint, not an official ATT&CK mapping.

Open ATT&CK lookup
cwe · low confidence lookup

CWE-770: Exact CWE lookup

Use the exact CWE identifier as the starting point before reviewing related ATT&CK behavior. Open the exact CWE lookup page first, then review the ATT&CK searches from that MITRE weakness context. This is a Glexia lookup hint, not an official ATT&CK mapping.

Open ATT&CK lookup
cve · low confidence lookup

CVE-2026-34756 mapping review

Open the CVE-to-ATT&CK bridge for reviewed, inferred, or future official mappings tied to this CVE.

Open ATT&CK lookup
Vulnerability profileCVE Program record
Severity
Medium
CVSS
6.5 (3.1)
Known Exploited
No
Published

Vector: CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H

Official CVE source material

CNA and ADP enrichment extracted from CVE v5

These fields come from the CVE record and ADP containers, not from Glexia's Take. They preserve time-varying source decisions such as CISA SSVC, KEV status, CVSS metrics, and provider references.

2CVSS vectors
5Timeline events
2ADP providers
15Source links

SSVC decision data

CISA-ADPCISA Coordinator
Timestamp
Version
2.0.3
Exploitation: noneAutomatable: noTechnical Impact: partial

CVSS vector scores

2 official scores

We collect every scored CVSS vector available in the official CNA and ADP containers. When more than one version is present, the table keeps the source vectors side by side instead of collapsing them into the highest score.

ScoreVersionSeverityVectorExploitImpactSource
6.5CVSS 3.1MediumCVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H2.83.6GitHub_M
6.5CVSS 3.1MediumCVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H2.83.6redhat-SADP

Vulnerability scoring details

Base CVSS 3.1 score

6.5Medium
CVSS 3.1 vector shape for CVE-2026-34756Attack VectorAttack ComplexityPrivileges RequiredUser InteractionScopeConfidentiality ImpactIntegrity ImpactAvailability Impact

Vector: CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H

Attack Vector
NetworkAdjacentLocalPhysical
Attack Complexity
LowHigh
Privileges Required
NoneLowHigh
User Interaction
NoneRequired
Scope
ChangedUnchanged
Confidentiality Impact
HighLowNone
Integrity Impact
HighLowNone
Availability Impact
HighLowNone

Vulnerability timeline

Timeline events are normalized from CVE metadata, CNA source timelines, ADP timelines, and KEV metadata when present.

  1. CVE reservedCVE Program

    The CVE ID was reserved by the assigning CNA.

  2. ADP timelineredhat-SADP

    Made public.

  3. CVE publishedCVE Program

    The CVE record was published.

  4. ADP timelineredhat-SADP

    Reported to Red Hat.

  5. CVE updatedCVE Program

    The CVE record metadata indicates this as the latest update time.

ADP provider summaries

CISA-ADPCISA ADP Vulnrichment
other:ssvc
redhat-SADPvllm: vLLM: Denial of Service via excessively large 'n' parameter in OpenAI-compatible API
other:Red Hat severity ratingcvssV3_1
  • 2026-04-06T16:03:45.222Z: Reported to Red Hat.
  • 2026-04-06T15:40:03.448Z: Made public.

Source materials

Affected products

Products and packages named in the record

VendorProductVersion / packageStatus
vllm-projectvllm>= 0.1.0, < 0.19.0Listed
Weakness

CWE details

CWE-1284CWE-770

CWE links open Glexia weakness intelligence pages with official CWE context, developer remediation guidance, and related CVE mappings.

CWE-1284 · source CWE mapping

Improper Validation of Specified Quantity in Input

Improper Validation of Specified Quantity in Input represents a recurring weakness pattern that can create exploitable paths when design, validation, or implementation controls are missing.

Open CWE intelligenceSame-CWE CVEs
CWE-770 · source CWE mapping

Allocation of Resources Without Limits or Throttling

Allocation of Resources Without Limits or Throttling represents a recurring weakness pattern that can create exploitable paths when design, validation, or implementation controls are missing.

Open CWE intelligenceSame-CWE CVEs