Security readout for executives and security teams
Plain-English summary
This vulnerability lets a logged-in attacker crash some vLLM servers by sending one oversized multimodal request containing many base64 JPEG frames. The issue is availability-only, but an outage in an LLM inference service can disrupt customer-facing AI features, internal automation, or dependent applications.
Executive priority
Treat as a timely service availability fix for exposed AI inference environments. Prioritize internet-facing, shared, or business-critical vLLM deployments first; lower urgency applies to isolated systems without multimodal video input.
Technical view
In vLLM 0.7.0 through versions before 0.19.0, VideoMediaIO.load_base64() handles video/jpeg data URLs by splitting comma-separated frames without enforcing num_frames. The load_bytes() path enforces the default frame limit, but this base64 path bypasses it, allowing excessive frame decoding into memory and possible OOM crash.
Likely exposure
Exposure is most likely where vLLM versions >=0.7.0 and <0.19.0 accept multimodal API requests with video/jpeg base64 content from authenticated or otherwise privileged users.
Exploitation context
The source bundle does not show KEV listing or cited active exploitation. CVSS indicates network access, low complexity, low privileges, no user interaction, and high availability impact, with no confidentiality or integrity impact.
Researcher notes
The key code-path distinction is base64 video/jpeg parsing versus byte-based loading. Evidence supports uncontrolled resource consumption through unbounded frame processing. Public sources name the fixed version but do not provide evidence of in-the-wild exploitation in this bundle.
Mitigation direction
- Upgrade vLLM to 0.19.0 or later.
- Apply relevant Red Hat errata where using Red Hat-packaged components.
- Review vendor guidance before relying on local workarounds.
- Restrict vLLM API access to trusted authenticated clients.
- Monitor and limit unusually large multimodal requests where supported.
Validation and detection
- Inventory deployed vLLM versions and flag >=0.7.0, <0.19.0.
- Confirm whether affected services accept video/jpeg base64 multimodal input.
- Verify vLLM is upgraded to 0.19.0 or vendor-fixed packaging.
- Review logs for abnormal large video/jpeg base64 requests and OOM crashes.
- Check Red Hat advisory status for any affected downstream products.
Public sources used
Generated from the cited source records. This long-tail analysis has not been individually reviewed by a named human.
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-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 lookupCVE-2026-34755 mapping review
Open the CVE-to-ATT&CK bridge for reviewed, inferred, or future official mappings tied to this CVE.
Open ATT&CK lookup- 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
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.
SSVC decision data
CVSS vector scores
2 official scoresWe 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.
CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H2.83.6GitHub_MCVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H2.83.6redhat-SADPVulnerability scoring details
Base CVSS 3.1 score
6.5MediumVector: CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H
Base CVSS 3.1 score
6.5MediumVector: CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H
Vulnerability timeline
Timeline events are normalized from CVE metadata, CNA source timelines, ADP timelines, and KEV metadata when present.
- CVE reservedCVE Program
The CVE ID was reserved by the assigning CNA.
- ADP timelineredhat-SADP
Made public.
- CVE publishedCVE Program
The CVE record was published.
- ADP timelineredhat-SADP
Reported to Red Hat.
- CVE updatedCVE Program
The CVE record metadata indicates this as the latest update time.
ADP provider summaries
- 2026-04-06T16:02:21.718Z: Reported to Red Hat.
- 2026-04-06T15:38:53.201Z: Made public.
Source materials
- CVE List V5 sourceCVE List V5
- https://github.com/vllm-project/vllm/security/advisories/GHSA-pq5c-rjhq-qp7pCVE reference, GitHub_M · x_refsource_CONFIRM
- https://access.redhat.com/security/cve/CVE-2026-34755CVE reference, redhat-SADP · vdb-entry, x_refsource_REDHAT
- RHBZ#2455403CVE reference, redhat-SADP · issue-tracking, x_refsource_REDHAT
- https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-34755.jsonCVE reference, redhat-SADP · x_sadp-csaf-vex
- https://access.redhat.com/errata/RHSA-2026:36005CVE reference, redhat-SADP · vendor-advisory, x_refsource_REDHAT
- https://access.redhat.com/errata/RHSA-2026:36006CVE reference, redhat-SADP · vendor-advisory, x_refsource_REDHAT
Products and packages named in the record
CWE details
CWE links open Glexia weakness intelligence pages with official CWE context, developer remediation guidance, and related CVE mappings.
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.