Skip to main content
Public proof of conceptFix available

CVE-2026-100650

A public proof of concept exists; no exploitation attested.

  • Published2026-09-26
  • WeaknessCWE-400

Am I affected?

With a domain, we read the version its public headers and pages expose. A distribution build can carry the fix under an older number.

Description

vLLM through 0.29.0 fetches and fully materializes remote or inline media before enforcing its documented media controls (the VLLM_MAX_AUDIO_CLIP_FILESIZE_MB compressed-audio size cap, default 25 MB, and the per-modality --limit-mm-per-prompt item limits). Across four ingress paths — the shared media-acquisition layer (HTTPConnection.get_bytes()/async_get_bytes()), the chat completions audio_url/base64 path, the batch speech runner, and the Rust frontend POST /tokenize route — the server reads the entire HTTP response body, base64-decodes the inline payload, or spawns one fetch/decode task per media part, and only then applies the limit (or, on some paths, never applies it). A remote attacker can therefore cause the API server or batch-runner process to allocate memory and consume outbound bandwidth proportional to an attacker-chosen body size or media item count before the request is rejected, resulting in pre-inference memory and bandwidth exhaustion (denial of service). The chat and batch surfaces require an API key when one is configured; the Rust frontend /tokenize route is unauthenticated by design. There is no code execution or data disclosure impact.

Timeline

dates in UTC
  1. 2026-09-2614:16CVE publishedNVD

Versions per branch

NVD · OSV
Vllm
BranchAffectedFixed in
0.29≤ 0.29.0no fix published
vllm
BranchAffectedFixed in
0.30< 0.30.00.30.0

as publishedRanges as NVD and OSV publish them; a distribution’s build can be fixed under an older number.

Products declared by the CNA1 product

As VulnCheck, the CNA that assigned this CVE, declared them in the CVE record; versions as written.

  • vllm-project vllmother versions: not affected
    Affected
    • < 0.29.0fixed in 0.29.0semver

    Not affected = 0.29.0

Source: the CNA record (CVE Program), © The MITRE Corporation,CVE Terms of UseCVE recordupdated 2026-10-01

Distribution builds and vendor advisories

Packages from OSV.dev and products from the vendors' CSAF advisories.

version not verifiedVersions not verified: read as OSV.dev and the vendors publish them, not checked against each vendor's release list.

Packages

  • vllmMinimOSall versions · no fixed version listedand 1 other package
  • vllm-cu129MinimOSall versions · fixed in 0.29.0-r0and 6 other packages

Vendor advisories

  • CVE-2026-100650 Red Hat Product Security39 products fixed · 11 known affected

    vllm: vLLM: Denial of Service via unbounded media ingestion

    rhaii/vllm-neuron-rhel9 as a component of Red Hat AI Inference Server · rhaii/vllm-tpu-rhel9 as a component of Red Hat AI Inference Server · rhaiis/vllm-cpu-rhel9 as a component of Red Hat AI Inference Server · +8 more

  • RHSA-2026:94738 Red Hat Product Security2 products fixed

    Red Hat Security Advisory: Red Hat AI Inference 3.5.3 (CUDA)

  • RHSA-2026:94721 Red Hat Product Security1 product fixed

    Red Hat Security Advisory: Red Hat AI Inference Tech Preview 3.5.3 (Gaudi)

  • RHSA-2026:94717 Red Hat Product Security3 products fixed

    Red Hat Security Advisory: Red Hat AI Inference 3.5.3 (Spyre)

  • RHSA-2026:94709 Red Hat Product Security1 product fixed

    Red Hat Security Advisory: Red Hat AI Inference 3.5.3 (ROCm)

  • RHSA-2026:94702 Red Hat Product Security1 product fixed

    Red Hat Security Advisory: Red Hat AI Inference 3.5.3 (CPU)

7 more advisories

Full record on NVD

Fix

  1. Vendor fix Red Hat Product Security RHSA-2026:94738 · 2 productsFor more information visit https://access.redhat.com/errata/RHSA-2026:94738
  2. Vendor fix Red Hat Product Security RHSA-2026:94721 · 1 productFor more information visit https://access.redhat.com/errata/RHSA-2026:94721
  3. Vendor fix Red Hat Product Security RHSA-2026:94717 · 3 productsFor more information visit https://access.redhat.com/errata/RHSA-2026:94717
  4. Vendor fix Red Hat Product Security RHSA-2026:94709 · 1 productFor more information visit https://access.redhat.com/errata/RHSA-2026:94709
  5. Vendor fix Red Hat Product Security RHSA-2026:94702 · 1 productFor more information visit https://access.redhat.com/errata/RHSA-2026:94702
  6. Vendor fix Red Hat Product Security RHSA-2026:90440 · 7 productsThe container images provided by this update can be downloaded from the Red Hat container registry at registry.redhat.io using the "podman pull" command. For details on deploying and configuring RHEL AI, see the Red Hat Enterprise Linux AI documentation at https://docs.redhat.com/en/documentation/…

Advisories and patches

Detection

Nuclei template
none so far
ProjectDiscovery
Exploit-DB
no entry
Exploit-DB
Proof of concept
Metasploit modules
none listed
Metasploit Framework

Sources

  • NVDCVE recordread 2026-10-23
  • CISA-ADPSSVC decisionpoc2026-09-30
  • CNACVSS 4.07.1known since 2026-10-23
  • CNACVSS 3.16.5known since 2026-10-23
  • CNAweakness (CWE)CWE-400known since 2026-10-23
  • FIRST EPSSEPSS score0.8 %2026-11-08
  • OSVaffected packages—
  • GitHub Advisoriesadvisory—

Identifiers

Technical details

CVSS v3 Vector
CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H
CVSS 4.0 vector
CVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X
Published
2026-09-26
Last Modified
2026-10-06
NVD status
Analyzed
GitHub Advisory
GHSA-j8gm-f3cq-6c27 GHSA-p6g9-7v3x-m8mv

Frequently asked questions

What is CVE-2026-100650?

vLLM through 0.29.0 fetches and fully materializes remote or inline media before enforcing its documented media controls (the VLLM_MAX_AUDIO_CLIP_FILESIZE_MB compressed-audio size cap, default 25 MB, and the per-modality --limit-mm-per-prompt item limits). Across four ingress paths — the shared media-acquisition layer (HTTPConnection.get_bytes()/async_get_bytes()), the chat completions audio_url/base64 path, the batch speech runner, and the Rust frontend POST /tokenize route — the server reads the entire HTTP response body, base64-decodes the inline payload, or spawns one fetch/decode task per media part, and only then applies the limit (or, on some paths, never applies it). A remote attacker can therefore cause the API server or batch-runner process to allocate memory and consume outbound bandwidth proportional to an attacker-chosen body size or media item count before the request is rejected, resulting in pre-inference memory and bandwidth exhaustion (denial of service). The chat and batch surfaces require an API key when one is configured; the Rust frontend /tokenize route is unauthenticated by design. There is no code execution or data disclosure impact.

Is CVE-2026-100650 actively exploited?

A proof-of-concept exploit exists for CVE-2026-100650, but active exploitation has not been confirmed at this time.

What is the CVSS score for CVE-2026-100650?

CVE-2026-100650 has a CVSS v3 base score of 6.5 (MEDIUM severity), with vector string CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H.

Running vllm? Hear about its next vulnerability

Surveillance reads the versions in your stack — a domain, a lockfile or a list — checks each CVE against the ranges its sources publish, and alerts you when one you run is exploited or enters CISA KEV. A version it cannot read stays « version not verified ».

Get alerts for this productCISA KEV additions, November 2026Search CVEs

No credit card required · No monthly cap on website lookups · 250 API requests/month

Go further

Related vulnerabilities

Same product first, then the same vendor advisory and the same publication week.

Back to CVE Database

Sources for this record