EUVD-2026-87744
GHSA-j8gm-f3cq-6c27
Severity by source
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
Network-reachable and low-complexity; PR:N because the Rust /tokenize ingress path is unauthenticated by design; impact is availability-only DoS, so C:N/I:N/A:H.
Primary rating from Vendor (VulnCheck).
CVSS VectorVendor: VulnCheck
Lifecycle Timeline
5Blast Radius
ecosystem impact- 3 pypi packages depend on vllm (3 direct, 0 indirect)
Ecosystem-wide dependent count for version 0.30.0.
DescriptionCVE.org
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.
Articles & Coverage 1
AnalysisAI
Pre-inference resource exhaustion in vLLM up to and including 0.29.0 lets remote attackers force the API server or batch runner to fully download, base64-decode and materialize attacker-controlled media before the documented VLLM_MAX_AUDIO_CLIP_FILESIZE_MB compressed-audio cap (25 MB default) and the per-modality --limit-mm-per-prompt item limits are enforced. Four ingress paths are affected - 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 - with the /tokenize path unauthenticated by design and the chat/batch surfaces gated only when an API key is actually configured, so an attacker can allocate memory and consume outbound bandwidth proportional to an attacker-chosen body size or media-item count. …
Unlock full vulnerability intelligence
- Risk assessment & exploitation conditions
- Attack chain visualization
- Remediation with exact patch versions
- Threat intelligence from 22 sources
- Personal watchlist & email alerts
No credit card · 7-day full trial
Attack ChainAIDerived
Hypothetical attack flow derived from CVE metadata
Vulnerability AssessmentAI
| Exploitation | Requires a running vLLM instance at or before 0.29.0 with a media-capable (multimodal) model serving one of 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, or the Rust frontend POST /tokenize route. … Additional conditions and limiting factors are described in the full assessment. |
| Risk Assessment | This is a genuine but bounded availability-only issue. … Full risk analysis with EPSS, KEV, and SSVC signal comparison available after sign-in. |
| Exploit Scenario | Full exploit scenario with step-by-step reproduction available after sign-in. |
| Remediation | Vendor-released patch: upgrade vLLM to the fixed release identified in the advisory (https://github.com/vllm-project/vllm/security/advisories/GHSA-p6g9-7v3x-m8mv), whose EUVD range indicates 0.29.0 as the corrected line - because the CVE text also says "through 0.29.0", confirm the exact fixed tag in the advisory and prefer the newest available release; the fix commit referenced is 752a3a504485790a2e8491cacbb35c137339ad34 (upstream fix available via commit/PR; release-level confirmation should come from the advisory). … Detailed patch versions, workarounds, and compensating controls in full report. |
Recommended ActionAI
Within 24 hours, inventory all vLLM instances and identify any running version 0.29.0 or earlier; immediately restrict public exposure of the API server, batch runner, and especially the /tokenize route using network controls, and enforce strict request body size and media-item count limits at the reverse proxy or API gateway. …
Sign in for detailed remediation steps and compensating controls.
Threat intelligence, references, and detailed analysis are available after sign-in.
Wazuh SIEM platform versions 4.4.0 through 4.9.0 contain an unsafe deserialization vulnerability in the DistributedAPI t
BentoML version 1.4.2 and earlier contains an unauthenticated remote code execution vulnerability through insecure deser
pgAdmin 4 contains critical remote code execution vulnerabilities in the Query Tool download and Cloud Deployment endpoi
The renderLocalView function in render/views.py in graphite-web in Graphite 0.9.5 through 0.9.10 uses the pickle Python
BentoML is a Python library for building online serving systems optimized for AI apps and model inference. Rated critica
OpenSSL before 0.9.8za, 1.0.0 before 1.0.0m, and 1.0.1 before 1.0.1h does not properly restrict processing of ChangeCiph
pyLoad download manager version prior to 0.5.0b3.dev77 exposes the Flask SECRET_KEY through an unauthenticated endpoint.
Langflow (a visual LLM pipeline builder) contains a critical unauthenticated code execution vulnerability (CVE-2026-3301
In Mercurial before 4.1.3, "hg serve --stdio" allows remote authenticated users to launch the Python debugger, and conse
Unauthenticated remote code execution affects Kestra OSS (the open-source event-driven orchestration platform) prior to
Unauthenticated remote code execution in Marimo ≤0.20.4 allows attackers to execute arbitrary system commands via the `/
pyLoad is the free and open-source Download Manager written in pure Python. Rated medium severity (CVSS 5.3), this vulne
Same weakness CWE-400 – Uncontrolled Resource Consumption
View allSame technique Denial Of Service
View allVendor StatusVendor
Share
External POC / Exploit Code
Leaving vuln.today