CVE-2026-90554 (GCVE-0-2026-90554)
Vulnerability from cvelistv5 – Published: 2026-09-12 12:08 – Updated: 2026-09-14 18:16
VLAI
EPSS
VEX
Title
vLLM before 0.28.0 Denial of Service via audio extraction
Summary
vLLM versions >=0.10.2 and <0.28.0 do not apply any audio decode-size or duration limit when extracting audio from video input for NanoNemotronVL models. In nano_nemotron_vl.py, _extract_audio_from_videos calls load_audio_pyav(BytesIO(video_bytes)) without the max_duration_s or max_decode_bytes parameters, so neither VLLM_MAX_AUDIO_DECODE_DURATION_S nor VLLM_MAX_AUDIO_DECODE_BYTES is enforced (unlike the direct audio upload path in AudioMediaIO). When a NanoNemotronVL model is served with use_audio_in_video=True, an attacker who supplies a small, highly compressed video as multimodal input can force the server to allocate gigabytes of memory during audio decoding, resulting in a denial of service. Fixed in vLLM 0.28.0.
Severity
6.2 (Medium)
SSVC
Exploitation: none
Automatable: no
Technical Impact: partial
CISA Coordinator · CISA-ADP (v2.0.3)
Decision recorded 2026-09-14 16:14 UTC
CWE
- CWE-400 - Uncontrolled Resource Consumption
Assigner
References
2 references
| URL | Tags |
|---|---|
| https://github.com/vllm-project/vllm/security/adv… | vendor-advisory |
| https://www.vulncheck.com/advisories/vllm-before-… | third-party-advisory |
Impacted products
1 product
| Vendor | Product | Version | CPE status | |
|---|---|---|---|---|
| vllm-project | vLLM |
Affected:
0.10.2 , < 0.28.0
(semver)
Unaffected: 0.28.0 (semver) cpe:2.3:a:vllm:vllm:*:*:*:*:*:*:*:* |
Date Public
2026-08-28 00:00
{
"containers": {
"adp": [
{
"metrics": [
{
"other": {
"content": {
"id": "CVE-2026-90554",
"options": [
{
"Exploitation": "none"
},
{
"Automatable": "no"
},
{
"Technical Impact": "partial"
}
],
"role": "CISA Coordinator",
"timestamp": "2026-09-14T16:14:37.760892Z",
"version": "2.0.3"
},
"type": "ssvc"
}
}
],
"providerMetadata": {
"dateUpdated": "2026-09-14T18:16:18.655Z",
"orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
"shortName": "CISA-ADP"
},
"title": "CISA ADP Vulnrichment"
}
],
"cna": {
"affected": [
{
"defaultStatus": "unaffected",
"packageURL": "pkg:pypi/vllm",
"product": "vLLM",
"vendor": "vllm-project",
"versions": [
{
"lessThan": "0.28.0",
"status": "affected",
"version": "0.10.2",
"versionType": "semver"
},
{
"status": "unaffected",
"version": "0.28.0",
"versionType": "semver"
}
]
}
],
"cpeApplicability": [
{
"nodes": [
{
"cpeMatch": [
{
"criteria": "cpe:2.3:a:vllm:vllm:*:*:*:*:*:*:*:*",
"versionEndExcluding": "0.28.0",
"versionStartIncluding": "0.10.2",
"vulnerable": true
}
],
"negate": false,
"operator": "OR"
}
]
}
],
"credits": [
{
"lang": "en",
"type": "reporter",
"value": "TobyB1702"
},
{
"lang": "en",
"type": "coordinator",
"value": "jperezdealgaba"
}
],
"datePublic": "2026-08-28T00:00:00.000Z",
"descriptions": [
{
"lang": "en",
"value": "vLLM versions \u003e=0.10.2 and \u003c0.28.0 do not apply any audio decode-size or duration limit when extracting audio from video input for NanoNemotronVL models. In nano_nemotron_vl.py, _extract_audio_from_videos calls load_audio_pyav(BytesIO(video_bytes)) without the max_duration_s or max_decode_bytes parameters, so neither VLLM_MAX_AUDIO_DECODE_DURATION_S nor VLLM_MAX_AUDIO_DECODE_BYTES is enforced (unlike the direct audio upload path in AudioMediaIO). When a NanoNemotronVL model is served with use_audio_in_video=True, an attacker who supplies a small, highly compressed video as multimodal input can force the server to allocate gigabytes of memory during audio decoding, resulting in a denial of service. Fixed in vLLM 0.28.0."
}
],
"metrics": [
{
"cvssV4_0": {
"attackComplexity": "LOW",
"attackRequirements": "NONE",
"attackVector": "LOCAL",
"baseScore": 6.9,
"baseSeverity": "MEDIUM",
"privilegesRequired": "NONE",
"subAvailabilityImpact": "NONE",
"subConfidentialityImpact": "NONE",
"subIntegrityImpact": "NONE",
"userInteraction": "NONE",
"vectorString": "CVSS:4.0/AV:L/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N",
"version": "4.0",
"vulnAvailabilityImpact": "HIGH",
"vulnConfidentialityImpact": "NONE",
"vulnIntegrityImpact": "NONE"
},
"format": "CVSS"
},
{
"cvssV3_1": {
"attackComplexity": "LOW",
"attackVector": "LOCAL",
"availabilityImpact": "HIGH",
"baseScore": 6.2,
"baseSeverity": "MEDIUM",
"confidentialityImpact": "NONE",
"integrityImpact": "NONE",
"privilegesRequired": "NONE",
"scope": "UNCHANGED",
"userInteraction": "NONE",
"vectorString": "CVSS:3.1/AV:L/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H",
"version": "3.1"
},
"format": "CVSS"
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-400",
"description": "Uncontrolled Resource Consumption",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-09-12T12:08:57.508Z",
"orgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"shortName": "VulnCheck"
},
"references": [
{
"name": "GitHub Security Advisory (GHSA-936p-m5pv-vvjf)",
"tags": [
"vendor-advisory"
],
"url": "https://github.com/vllm-project/vllm/security/advisories/GHSA-936p-m5pv-vvjf"
},
{
"name": "VulnCheck Advisory: vLLM before 0.28.0 Denial of Service via audio extraction",
"tags": [
"third-party-advisory"
],
"url": "https://www.vulncheck.com/advisories/vllm-before-0.28.0-denial-of-service-via-audio-extraction"
}
],
"title": "vLLM before 0.28.0 Denial of Service via audio extraction",
"x_generator": {
"engine": "vulncheck-endgame"
}
}
},
"cveMetadata": {
"assignerOrgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"assignerShortName": "VulnCheck",
"cveId": "CVE-2026-90554",
"datePublished": "2026-09-12T12:08:57.508Z",
"dateReserved": "2026-09-12T11:13:43.326Z",
"dateUpdated": "2026-09-14T18:16:18.655Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2",
"vulnerability-lookup:meta": {
"epss": {
"cve": "CVE-2026-90554",
"date": "2026-10-27",
"epss": "0.00201",
"percentile": "0.09848"
},
"nvd": {
"cve": {
"affected": [
{
"affectedData": [
{
"defaultStatus": "unaffected",
"packageURL": "pkg:pypi/vllm",
"product": "vLLM",
"vendor": "vllm-project",
"versions": [
{
"lessThan": "0.28.0",
"status": "affected",
"version": "0.10.2",
"versionType": "semver"
},
{
"status": "unaffected",
"version": "0.28.0",
"versionType": "semver"
}
]
}
],
"source": "disclosure@vulncheck.com"
}
],
"configurations": [
{
"nodes": [
{
"cpeMatch": [
{
"criteria": "cpe:2.3:a:vllm:vllm:*:*:*:*:*:*:*:*",
"matchCriteriaId": "BF7219AF-B221-49B8-AFBC-03DB4E2FAD52",
"versionEndExcluding": "0.28.0",
"versionStartIncluding": "0.10.2",
"vulnerable": true
}
],
"negate": false,
"operator": "OR"
}
]
}
],
"cveTags": [],
"descriptions": [
{
"lang": "en",
"value": "vLLM versions \u003e=0.10.2 and \u003c0.28.0 do not apply any audio decode-size or duration limit when extracting audio from video input for NanoNemotronVL models. In nano_nemotron_vl.py, _extract_audio_from_videos calls load_audio_pyav(BytesIO(video_bytes)) without the max_duration_s or max_decode_bytes parameters, so neither VLLM_MAX_AUDIO_DECODE_DURATION_S nor VLLM_MAX_AUDIO_DECODE_BYTES is enforced (unlike the direct audio upload path in AudioMediaIO). When a NanoNemotronVL model is served with use_audio_in_video=True, an attacker who supplies a small, highly compressed video as multimodal input can force the server to allocate gigabytes of memory during audio decoding, resulting in a denial of service. Fixed in vLLM 0.28.0."
}
],
"id": "CVE-2026-90554",
"lastModified": "2026-09-24T20:28:13.590",
"metrics": {
"cvssMetricV31": [
{
"cvssData": {
"attackComplexity": "LOW",
"attackVector": "LOCAL",
"availabilityImpact": "HIGH",
"baseScore": 6.2,
"baseSeverity": "MEDIUM",
"confidentialityImpact": "NONE",
"integrityImpact": "NONE",
"privilegesRequired": "NONE",
"scope": "UNCHANGED",
"userInteraction": "NONE",
"vectorString": "CVSS:3.1/AV:L/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H",
"version": "3.1"
},
"exploitabilityScore": 2.5,
"impactScore": 3.6,
"source": "disclosure@vulncheck.com",
"type": "Secondary"
},
{
"cvssData": {
"attackComplexity": "LOW",
"attackVector": "NETWORK",
"availabilityImpact": "HIGH",
"baseScore": 7.5,
"baseSeverity": "HIGH",
"confidentialityImpact": "NONE",
"integrityImpact": "NONE",
"privilegesRequired": "NONE",
"scope": "UNCHANGED",
"userInteraction": "NONE",
"vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H",
"version": "3.1"
},
"exploitabilityScore": 3.9,
"impactScore": 3.6,
"source": "nvd@nist.gov",
"type": "Primary"
}
],
"cvssMetricV40": [
{
"cvssData": {
"Automatable": "NOT_DEFINED",
"Recovery": "NOT_DEFINED",
"Safety": "NOT_DEFINED",
"attackComplexity": "LOW",
"attackRequirements": "NONE",
"attackVector": "LOCAL",
"availabilityRequirement": "NOT_DEFINED",
"baseScore": 6.9,
"baseSeverity": "MEDIUM",
"confidentialityRequirement": "NOT_DEFINED",
"exploitMaturity": "NOT_DEFINED",
"integrityRequirement": "NOT_DEFINED",
"modifiedAttackComplexity": "NOT_DEFINED",
"modifiedAttackRequirements": "NOT_DEFINED",
"modifiedAttackVector": "NOT_DEFINED",
"modifiedPrivilegesRequired": "NOT_DEFINED",
"modifiedSubAvailabilityImpact": "NOT_DEFINED",
"modifiedSubConfidentialityImpact": "NOT_DEFINED",
"modifiedSubIntegrityImpact": "NOT_DEFINED",
"modifiedUserInteraction": "NOT_DEFINED",
"modifiedVulnAvailabilityImpact": "NOT_DEFINED",
"modifiedVulnConfidentialityImpact": "NOT_DEFINED",
"modifiedVulnIntegrityImpact": "NOT_DEFINED",
"privilegesRequired": "NONE",
"providerUrgency": "NOT_DEFINED",
"subAvailabilityImpact": "NONE",
"subConfidentialityImpact": "NONE",
"subIntegrityImpact": "NONE",
"userInteraction": "NONE",
"valueDensity": "NOT_DEFINED",
"vectorString": "CVSS:4.0/AV:L/AC:L/AT:N/PR:N/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",
"version": "4.0",
"vulnAvailabilityImpact": "HIGH",
"vulnConfidentialityImpact": "NONE",
"vulnIntegrityImpact": "NONE",
"vulnerabilityResponseEffort": "NOT_DEFINED"
},
"source": "disclosure@vulncheck.com",
"type": "Secondary"
}
],
"ssvcV203": [
{
"source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
"ssvcData": {
"id": "CVE-2026-90554",
"options": [
{
"exploitation": "none"
},
{
"automatable": "no"
},
{
"technicalImpact": "partial"
}
],
"role": "CISA Coordinator",
"timestamp": "2026-09-14T16:14:37.760892Z",
"version": "2.0.3"
}
}
]
},
"published": "2026-09-12T13:16:54.040",
"references": [
{
"source": "disclosure@vulncheck.com",
"tags": [
"Vendor Advisory"
],
"url": "https://github.com/vllm-project/vllm/security/advisories/GHSA-936p-m5pv-vvjf"
},
{
"source": "disclosure@vulncheck.com",
"tags": [
"Third Party Advisory"
],
"url": "https://www.vulncheck.com/advisories/vllm-before-0.28.0-denial-of-service-via-audio-extraction"
}
],
"sourceIdentifier": "disclosure@vulncheck.com",
"vulnStatus": "Analyzed",
"weaknesses": [
{
"description": [
{
"lang": "en",
"value": "CWE-400"
}
],
"source": "disclosure@vulncheck.com",
"type": "Secondary"
}
]
}
},
"redhat_vex": {
"aggregate_severity": "Moderate",
"current_release_date": "2026-09-24T20:51:41+00:00",
"cve": "CVE-2026-90554",
"id": "CVE-2026-90554",
"initial_release_date": "2026-09-12T12:08:57.508000+00:00",
"product_status:known_affected": "27",
"source": "Red Hat CSAF VEX",
"status": "final",
"title": "vllm: vLLM: Denial of Service via video audio extraction",
"url": "https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-90554.json",
"version": "3"
},
"vulnrichment": {
"containers": {
"adp": [
{
"metrics": [
{
"other": {
"content": {
"id": "CVE-2026-90554",
"options": [
{
"Exploitation": "none"
},
{
"Automatable": "no"
},
{
"Technical Impact": "partial"
}
],
"role": "CISA Coordinator",
"timestamp": "2026-09-14T16:14:37.760892Z",
"version": "2.0.3"
},
"type": "ssvc"
}
}
],
"providerMetadata": {
"dateUpdated": "2026-09-14T16:14:38.770Z",
"orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
"shortName": "CISA-ADP"
},
"title": "CISA ADP Vulnrichment"
}
],
"cna": {
"affected": [
{
"defaultStatus": "unaffected",
"packageURL": "pkg:pypi/vllm",
"product": "vLLM",
"vendor": "vllm-project",
"versions": [
{
"lessThan": "0.28.0",
"status": "affected",
"version": "0.10.2",
"versionType": "semver"
},
{
"status": "unaffected",
"version": "0.28.0",
"versionType": "semver"
}
]
}
],
"cpeApplicability": [
{
"nodes": [
{
"cpeMatch": [
{
"criteria": "cpe:2.3:a:vllm:vllm:*:*:*:*:*:*:*:*",
"versionEndExcluding": "0.28.0",
"versionStartIncluding": "0.10.2",
"vulnerable": true
}
],
"negate": false,
"operator": "OR"
}
]
}
],
"credits": [
{
"lang": "en",
"type": "reporter",
"value": "TobyB1702"
},
{
"lang": "en",
"type": "coordinator",
"value": "jperezdealgaba"
}
],
"datePublic": "2026-08-28T00:00:00.000Z",
"descriptions": [
{
"lang": "en",
"value": "vLLM versions \u003e=0.10.2 and \u003c0.28.0 do not apply any audio decode-size or duration limit when extracting audio from video input for NanoNemotronVL models. In nano_nemotron_vl.py, _extract_audio_from_videos calls load_audio_pyav(BytesIO(video_bytes)) without the max_duration_s or max_decode_bytes parameters, so neither VLLM_MAX_AUDIO_DECODE_DURATION_S nor VLLM_MAX_AUDIO_DECODE_BYTES is enforced (unlike the direct audio upload path in AudioMediaIO). When a NanoNemotronVL model is served with use_audio_in_video=True, an attacker who supplies a small, highly compressed video as multimodal input can force the server to allocate gigabytes of memory during audio decoding, resulting in a denial of service. Fixed in vLLM 0.28.0."
}
],
"metrics": [
{
"cvssV4_0": {
"attackComplexity": "LOW",
"attackRequirements": "NONE",
"attackVector": "LOCAL",
"baseScore": 6.9,
"baseSeverity": "MEDIUM",
"privilegesRequired": "NONE",
"subAvailabilityImpact": "NONE",
"subConfidentialityImpact": "NONE",
"subIntegrityImpact": "NONE",
"userInteraction": "NONE",
"vectorString": "CVSS:4.0/AV:L/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N",
"version": "4.0",
"vulnAvailabilityImpact": "HIGH",
"vulnConfidentialityImpact": "NONE",
"vulnIntegrityImpact": "NONE"
},
"format": "CVSS"
},
{
"cvssV3_1": {
"attackComplexity": "LOW",
"attackVector": "LOCAL",
"availabilityImpact": "HIGH",
"baseScore": 6.2,
"baseSeverity": "MEDIUM",
"confidentialityImpact": "NONE",
"integrityImpact": "NONE",
"privilegesRequired": "NONE",
"scope": "UNCHANGED",
"userInteraction": "NONE",
"vectorString": "CVSS:3.1/AV:L/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H",
"version": "3.1"
},
"format": "CVSS"
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-400",
"description": "Uncontrolled Resource Consumption",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-09-12T12:08:57.508Z",
"orgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"shortName": "VulnCheck"
},
"references": [
{
"name": "GitHub Security Advisory (GHSA-936p-m5pv-vvjf)",
"tags": [
"vendor-advisory"
],
"url": "https://github.com/vllm-project/vllm/security/advisories/GHSA-936p-m5pv-vvjf"
},
{
"name": "VulnCheck Advisory: vLLM before 0.28.0 Denial of Service via audio extraction",
"tags": [
"third-party-advisory"
],
"url": "https://www.vulncheck.com/advisories/vllm-before-0.28.0-denial-of-service-via-audio-extraction"
}
],
"title": "vLLM before 0.28.0 Denial of Service via audio extraction",
"x_generator": {
"engine": "vulncheck-endgame"
}
}
},
"cveMetadata": {
"assignerOrgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"assignerShortName": "VulnCheck",
"cveId": "CVE-2026-90554",
"datePublished": "2026-09-12T12:08:57.508Z",
"dateReserved": "2026-09-12T11:13:43.326Z",
"dateUpdated": "2026-09-14T18:16:18.655Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}
}
}
Loading…
Loading…
Drag nodes to arrange them. Double-click a related vulnerability to open it, or a source to visit its link. The Physics button on the left tunes the layout and the simulation; the legend at the bottom left hides or shows nodes by kind and by sighting type.
Experimental. This forecast is provided for visualization only and may change without notice. Do not use it for operational decisions.
Forecast uses a logistic model when the trend is rising, or an exponential decay model when the trend is falling. Fitted via linearized least squares.
Sightings
| Author | Source | Type | Date | Other |
|---|
Nomenclature
- Seen: The vulnerability was mentioned, discussed, or observed by the user.
- Confirmed: The vulnerability has been validated from an analyst's perspective.
- Published Proof of Concept: A public proof of concept is available for this vulnerability.
- Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
- Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
- Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
- Not confirmed: The user expressed doubt about the validity of the vulnerability.
- Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.
Vulnerability timeline
Sightings and recorded dates, grouped by day · UTC
Loading timeline…
Loading…
Loading…
The MITRE ATT&CK techniques below are AI-generated suggestions, inferred from the description of the
vulnerability by the CIRCL/vulnerability-attack-technique-classification-roberta-base
model, served locally by ML-Gateway.
They have not been verified by an analyst and are provided for guidance only.
The approach is described in our paper Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion.
Browse all ATT&CK techniques and the vulnerabilities related to each.
The approach is described in our paper Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion.
Browse all ATT&CK techniques and the vulnerabilities related to each.
Loading…
Related by attack behaviour
Vulnerabilities whose description is nearest to this one in the vector space of the CIRCL/vulnerability-attack-technique-biencoder model. This is a similarity search over the bi-encoder space (plain cosine), not a classification, and it has no measured accuracy.