CVE-2026-55646

{"dataType": "CVE_RECORD", "dataVersion": "5.2", "cveMetadata": {"cveId": "CVE-2026-55646", "assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa", "state": "PUBLISHED", "assignerShortName": "GitHub_M", "dateReserved": "2026-06-16T23:52:12.058Z", "datePublished": "2026-07-06T19:41:11.330Z", "dateUpdated": "2026-07-06T19:49:44.773Z"}, "containers": {"cna": {"title": "vLLM speech-to-text endpoints allocate full upload before enforcing the audio file-size limit", "problemTypes": [{"descriptions": [{"cweId": "CWE-400", "lang": "en", "description": "CWE-400: Uncontrolled Resource Consumption", "type": "CWE"}]}, {"descriptions": [{"cweId": "CWE-770", "lang": "en", "description": "CWE-770: Allocation of Resources Without Limits or Throttling", "type": "CWE"}]}], "metrics": [{"cvssV3_1": {"attackComplexity": "LOW", "attackVector": "NETWORK", "availabilityImpact": "HIGH", "baseScore": 6.5, "baseSeverity": "MEDIUM", "confidentialityImpact": "NONE", "integrityImpact": "NONE", "privilegesRequired": "LOW", "scope": "UNCHANGED", "userInteraction": "NONE", "vectorString": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H", "version": "3.1"}}], "references": [{"name": "https://github.com/vllm-project/vllm/security/advisories/GHSA-v82g-2437-67m2", "tags": ["x_refsource_CONFIRM"], "url": "https://github.com/vllm-project/vllm/security/advisories/GHSA-v82g-2437-67m2"}, {"name": "https://github.com/vllm-project/vllm/pull/45510", "tags": ["x_refsource_MISC"], "url": "https://github.com/vllm-project/vllm/pull/45510"}, {"name": "https://github.com/vllm-project/vllm/commit/b997071ec493765abbed990c65843ed05e4708a8", "tags": ["x_refsource_MISC"], "url": "https://github.com/vllm-project/vllm/commit/b997071ec493765abbed990c65843ed05e4708a8"}], "affected": [{"vendor": "vllm-project", "product": "vllm", "versions": [{"version": ">= 0.22.0, < 0.24.0", "status": "affected"}]}], "providerMetadata": {"orgId": "a0819718-46f1-4df5-94e2-005712e83aaa", "shortName": "GitHub_M", "dateUpdated": "2026-07-06T19:41:11.330Z"}, "descriptions": [{"lang": "en", "value": "vLLM is an inference and serving engine for large language models. From 0.22.0 to 0.23.0, the /v1/audio/transcriptions and /v1/audio/translations routes call request.file.read() to fully materialize an uploaded audio file into memory before vLLM checks the documented VLLM_MAX_AUDIO_CLIP_FILESIZE_MB compressed upload size limit (default 25 MB) later in the speech-to-text preprocessing step, so an API caller who can reach those routes can submit an oversized multipart upload and cause vLLM to allocate memory proportional to the uploaded file size before the request is rejected as too large, creating memory pressure or terminating the process depending on deployment resource limits. This issue is fixed in version 0.24.0."}], "source": {"advisory": "GHSA-v82g-2437-67m2", "discovery": "UNKNOWN"}}, "adp": [{"references": [{"url": "https://github.com/vllm-project/vllm/security/advisories/GHSA-v82g-2437-67m2", "tags": ["exploit"]}], "metrics": [{"other": {"type": "ssvc", "content": {"timestamp": "2026-07-06T19:49:41.180992Z", "id": "CVE-2026-55646", "options": [{"Exploitation": "poc"}, {"Automatable": "no"}, {"Technical Impact": "partial"}], "role": "CISA Coordinator", "version": "2.0.3"}}}], "title": "CISA ADP Vulnrichment", "providerMetadata": {"orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0", "shortName": "CISA-ADP", "dateUpdated": "2026-07-06T19:49:44.773Z"}}]}}

{"dataType": "CVE_RECORD", "containers": {"adp": [{"title": "CISA ADP Vulnrichment", "metrics": [{"other": {"type": "ssvc", "content": {"id": "CVE-2026-55646", "role": "CISA Coordinator", "options": [{"Exploitation": "poc"}, {"Automatable": "no"}, {"Technical Impact": "partial"}], "version": "2.0.3", "timestamp": "2026-07-06T19:49:41.180992Z"}}}], "references": [{"url": "https://github.com/vllm-project/vllm/security/advisories/GHSA-v82g-2437-67m2", "tags": ["exploit"]}], "providerMetadata": {"orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0", "shortName": "CISA-ADP", "dateUpdated": "2026-07-06T19:49:27.923Z"}}], "cna": {"title": "vLLM speech-to-text endpoints allocate full upload before enforcing the audio file-size limit", "source": {"advisory": "GHSA-v82g-2437-67m2", "discovery": "UNKNOWN"}, "metrics": [{"cvssV3_1": {"scope": "UNCHANGED", "version": "3.1", "baseScore": 6.5, "attackVector": "NETWORK", "baseSeverity": "MEDIUM", "vectorString": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H", "integrityImpact": "NONE", "userInteraction": "NONE", "attackComplexity": "LOW", "availabilityImpact": "HIGH", "privilegesRequired": "LOW", "confidentialityImpact": "NONE"}}], "affected": [{"vendor": "vllm-project", "product": "vllm", "versions": [{"status": "affected", "version": ">= 0.22.0, < 0.24.0"}]}], "references": [{"url": "https://github.com/vllm-project/vllm/security/advisories/GHSA-v82g-2437-67m2", "name": "https://github.com/vllm-project/vllm/security/advisories/GHSA-v82g-2437-67m2", "tags": ["x_refsource_CONFIRM"]}, {"url": "https://github.com/vllm-project/vllm/pull/45510", "name": "https://github.com/vllm-project/vllm/pull/45510", "tags": ["x_refsource_MISC"]}, {"url": "https://github.com/vllm-project/vllm/commit/b997071ec493765abbed990c65843ed05e4708a8", "name": "https://github.com/vllm-project/vllm/commit/b997071ec493765abbed990c65843ed05e4708a8", "tags": ["x_refsource_MISC"]}], "descriptions": [{"lang": "en", "value": "vLLM is an inference and serving engine for large language models. From 0.22.0 to 0.23.0, the /v1/audio/transcriptions and /v1/audio/translations routes call request.file.read() to fully materialize an uploaded audio file into memory before vLLM checks the documented VLLM_MAX_AUDIO_CLIP_FILESIZE_MB compressed upload size limit (default 25 MB) later in the speech-to-text preprocessing step, so an API caller who can reach those routes can submit an oversized multipart upload and cause vLLM to allocate memory proportional to the uploaded file size before the request is rejected as too large, creating memory pressure or terminating the process depending on deployment resource limits. This issue is fixed in version 0.24.0."}], "problemTypes": [{"descriptions": [{"lang": "en", "type": "CWE", "cweId": "CWE-400", "description": "CWE-400: Uncontrolled Resource Consumption"}]}, {"descriptions": [{"lang": "en", "type": "CWE", "cweId": "CWE-770", "description": "CWE-770: Allocation of Resources Without Limits or Throttling"}]}], "providerMetadata": {"orgId": "a0819718-46f1-4df5-94e2-005712e83aaa", "shortName": "GitHub_M", "dateUpdated": "2026-07-06T19:41:11.330Z"}}}, "cveMetadata": {"cveId": "CVE-2026-55646", "state": "PUBLISHED", "dateUpdated": "2026-07-06T19:49:44.773Z", "dateReserved": "2026-06-16T23:52:12.058Z", "assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa", "datePublished": "2026-07-06T19:41:11.330Z", "assignerShortName": "GitHub_M"}, "dataVersion": "5.2"}

{"affected": [{"affectedData": [{"product": "vllm", "vendor": "vllm-project", "versions": [{"status": "affected", "version": ">= 0.22.0, < 0.24.0"}]}], "source": "security-advisories@github.com"}], "configurations": [{"nodes": [{"cpeMatch": [{"criteria": "cpe:2.3:a:vllm:vllm:*:*:*:*:*:*:*:*", "matchCriteriaId": "D7860DB8-F910-4CD7-92D1-4775DA105A7F", "versionEndExcluding": "0.24.0", "versionStartIncluding": "0.22.0", "vulnerable": true}], "negate": false, "operator": "OR"}]}], "cveTags": [], "descriptions": [{"lang": "en", "value": "vLLM is an inference and serving engine for large language models. From 0.22.0 to 0.23.0, the /v1/audio/transcriptions and /v1/audio/translations routes call request.file.read() to fully materialize an uploaded audio file into memory before vLLM checks the documented VLLM_MAX_AUDIO_CLIP_FILESIZE_MB compressed upload size limit (default 25 MB) later in the speech-to-text preprocessing step, so an API caller who can reach those routes can submit an oversized multipart upload and cause vLLM to allocate memory proportional to the uploaded file size before the request is rejected as too large, creating memory pressure or terminating the process depending on deployment resource limits. This issue is fixed in version 0.24.0."}, {"lang": "es", "value": "vLLM es un motor de inferencia y servicio para modelos de lenguaje grandes. Desde la versi\u00f3n 0.22.0 hasta la 0.23.0, las rutas /v1/audio/transcriptions y /v1/audio/translations llaman a request.file.read() para materializar completamente un archivo de audio subido en la memoria antes de que vLLM verifique el l\u00edmite documentado de tama\u00f1o de carga comprimida VLLM_MAX_AUDIO_CLIP_FILESIZE_MB (predeterminado 25 MB) m\u00e1s tarde en el paso de preprocesamiento de voz a texto, por lo que un llamador de API que pueda alcanzar esas rutas puede enviar una carga multipart sobredimensionada y hacer que vLLM asigne memoria proporcional al tama\u00f1o del archivo subido antes de que la solicitud sea rechazada por ser demasiado grande, creando presi\u00f3n de memoria o terminando el proceso dependiendo de los l\u00edmites de recursos de despliegue. Este problema est\u00e1 solucionado en la versi\u00f3n 0.24.0."}], "id": "CVE-2026-55646", "lastModified": "2026-10-25T20:10:00.147", "metrics": {"cvssMetricV31": [{"cvssData": {"attackComplexity": "LOW", "attackVector": "NETWORK", "availabilityImpact": "HIGH", "baseScore": 6.5, "baseSeverity": "MEDIUM", "confidentialityImpact": "NONE", "integrityImpact": "NONE", "privilegesRequired": "LOW", "scope": "UNCHANGED", "userInteraction": "NONE", "vectorString": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H", "version": "3.1"}, "exploitabilityScore": 2.8, "impactScore": 3.6, "source": "security-advisories@github.com", "type": "Secondary"}], "ssvcV203": [{"source": "134c704f-9b21-4f2e-91b3-4a467353bcc0", "ssvcData": {"id": "CVE-2026-55646", "options": [{"exploitation": "poc"}, {"automatable": "no"}, {"technicalImpact": "partial"}], "role": "CISA Coordinator", "timestamp": "2026-07-06T19:49:41.180992Z", "version": "2.0.3"}}]}, "published": "2026-07-06T20:16:37.663", "references": [{"source": "security-advisories@github.com", "tags": ["Patch"], "url": "https://github.com/vllm-project/vllm/commit/b997071ec493765abbed990c65843ed05e4708a8"}, {"source": "security-advisories@github.com", "tags": ["Issue Tracking", "Patch"], "url": "https://github.com/vllm-project/vllm/pull/45510"}, {"source": "security-advisories@github.com", "tags": ["Vendor Advisory"], "url": "https://github.com/vllm-project/vllm/security/advisories/GHSA-v82g-2437-67m2"}, {"source": "134c704f-9b21-4f2e-91b3-4a467353bcc0", "tags": ["Vendor Advisory"], "url": "https://github.com/vllm-project/vllm/security/advisories/GHSA-v82g-2437-67m2"}], "sourceIdentifier": "security-advisories@github.com", "vulnStatus": "Analyzed", "weaknesses": [{"description": [{"lang": "en", "value": "CWE-400"}, {"lang": "en", "value": "CWE-770"}], "source": "security-advisories@github.com", "type": "Secondary"}]}

{"bugzilla": {"description": "vllm: vLLM: Denial of Service due to excessive memory allocation via oversized audio file uploads", "id": "2497468", "url": "https://bugzilla.redhat.com/show_bug.cgi?id=2497468"}, "csaw": false, "cvss3": {"cvss3_base_score": "6.5", "cvss3_scoring_vector": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H", "status": "draft"}, "cwe": "CWE-770", "details": ["vLLM is an inference and serving engine for large language models. From 0.22.0 to 0.23.0, the /v1/audio/transcriptions and /v1/audio/translations routes call request.file.read() to fully materialize an uploaded audio file into memory before vLLM checks the documented VLLM_MAX_AUDIO_CLIP_FILESIZE_MB compressed upload size limit (default 25 MB) later in the speech-to-text preprocessing step, so an API caller who can reach those routes can submit an oversized multipart upload and cause vLLM to allocate memory proportional to the uploaded file size before the request is rejected as too large, creating memory pressure or terminating the process depending on deployment resource limits. This issue is fixed in version 0.24.0.", "A flaw was found in vLLM, an inference and serving engine for large language models. A remote attacker, by submitting an oversized audio file to the `/v1/audio/transcriptions` or `/v1/audio/translations` routes, can cause the system to allocate excessive memory. This occurs before the configured file size limit is checked, leading to memory pressure or process termination, resulting in a Denial of Service (DoS)."], "mitigation": {"lang": "en:us", "value": "Mitigation for this issue is either not available or the currently available options do not meet the Red Hat Product Security criteria comprising ease of use and deployment, applicability to widespread installation base, or stability."}, "name": "CVE-2026-55646", "package_state": [{"cpe": "cpe:/a:redhat:ai_inference_server:3", "fix_state": "Not affected", "package_name": "rhaiis/vllm-cpu-rhel9", "product_name": "Red Hat AI Inference Server"}, {"cpe": "cpe:/a:redhat:ai_inference_server:3", "fix_state": "Not affected", "package_name": "rhaiis/vllm-cuda-rhel9", "product_name": "Red Hat AI Inference Server"}, {"cpe": "cpe:/a:redhat:ai_inference_server:3", "fix_state": "Not affected", "package_name": "rhaiis/vllm-neuron-rhel9", "product_name": "Red Hat AI Inference Server"}, {"cpe": "cpe:/a:redhat:ai_inference_server:3", "fix_state": "Not affected", "package_name": "rhaiis/vllm-rocm-rhel9", "product_name": "Red Hat AI Inference Server"}, {"cpe": "cpe:/a:redhat:ai_inference_server:3", "fix_state": "Not affected", "package_name": "rhaiis/vllm-spyre-rhel9", "product_name": "Red Hat AI Inference Server"}, {"cpe": "cpe:/a:redhat:ai_inference_server:3", "fix_state": "Not affected", "package_name": "rhaiis/vllm-tpu-rhel9", "product_name": "Red Hat AI Inference Server"}, {"cpe": "cpe:/a:redhat:ai_inference_server:3", "fix_state": "Not affected", "package_name": "rhaii/vllm-cpu-rhel9", "product_name": "Red Hat AI Inference Server"}, {"cpe": "cpe:/a:redhat:ai_inference_server:3", "fix_state": "Not affected", "package_name": "rhaii/vllm-cuda-rhel9", "product_name": "Red Hat AI Inference Server"}, {"cpe": "cpe:/a:redhat:ai_inference_server:3", "fix_state": "Not affected", "package_name": "rhaii/vllm-gaudi-rhel9", "product_name": "Red Hat AI Inference Server"}, {"cpe": "cpe:/a:redhat:ai_inference_server:3", "fix_state": "Not affected", "package_name": "rhaii/vllm-neuron-rhel9", "product_name": "Red Hat AI Inference Server"}, {"cpe": "cpe:/a:redhat:ai_inference_server:3", "fix_state": "Not affected", "package_name": "rhaii/vllm-rocm-rhel9", "product_name": "Red Hat AI Inference Server"}, {"cpe": "cpe:/a:redhat:ai_inference_server:3", "fix_state": "Not affected", "package_name": "rhaii/vllm-spyre-rhel9", "product_name": "Red Hat AI Inference Server"}, {"cpe": "cpe:/a:redhat:ai_inference_server:3", "fix_state": "Not affected", "package_name": "rhaii/vllm-tpu-rhel9", "product_name": "Red Hat AI Inference Server"}, {"cpe": "cpe:/a:redhat:enterprise_linux_ai:3", "fix_state": "Not affected", "package_name": "rhelai3/bootc-aws-cuda-rhel9", "product_name": "Red Hat Enterprise Linux AI (RHEL AI) 3"}, {"cpe": "cpe:/a:redhat:enterprise_linux_ai:3", "fix_state": "Not affected", "package_name": "rhelai3/bootc-azure-cuda-rhel9", "product_name": "Red Hat Enterprise Linux AI (RHEL AI) 3"}, {"cpe": "cpe:/a:redhat:enterprise_linux_ai:3", "fix_state": "Not affected", "package_name": "rhelai3/bootc-azure-rocm-rhel9", "product_name": "Red Hat Enterprise Linux AI (RHEL AI) 3"}, {"cpe": "cpe:/a:redhat:enterprise_linux_ai:3", "fix_state": "Not affected", "package_name": "rhelai3/bootc-cuda-rhel9", "product_name": "Red Hat Enterprise Linux AI (RHEL AI) 3"}, {"cpe": "cpe:/a:redhat:enterprise_linux_ai:3", "fix_state": "Not affected", "package_name": "rhelai3/bootc-gaudi-rhel9", "product_name": "Red Hat Enterprise Linux AI (RHEL AI) 3"}, {"cpe": "cpe:/a:redhat:enterprise_linux_ai:3", "fix_state": "Not affected", "package_name": "rhelai3/bootc-gcp-cuda-rhel9", "product_name": "Red Hat Enterprise Linux AI (RHEL AI) 3"}, {"cpe": "cpe:/a:redhat:enterprise_linux_ai:3", "fix_state": "Not affected", "package_name": "rhelai3/bootc-rocm-rhel9", "product_name": "Red Hat Enterprise Linux AI (RHEL AI) 3"}, {"cpe": "cpe:/a:redhat:openshift_ai", "fix_state": "Not affected", "package_name": "rhoai/odh-kserve-agent-rhel9", "product_name": "Red Hat OpenShift AI (RHOAI)"}, {"cpe": "cpe:/a:redhat:openshift_ai", "fix_state": "Not affected", "package_name": "rhoai/odh-kserve-controller-rhel9", "product_name": "Red Hat OpenShift AI (RHOAI)"}, {"cpe": "cpe:/a:redhat:openshift_ai", "fix_state": "Not affected", "package_name": "rhoai/odh-kserve-router-rhel9", "product_name": "Red Hat OpenShift AI (RHOAI)"}, {"cpe": "cpe:/a:redhat:openshift_ai", "fix_state": "Not affected", "package_name": "rhoai/odh-kserve-storage-initializer-rhel9", "product_name": "Red Hat OpenShift AI (RHOAI)"}, {"cpe": "cpe:/a:redhat:openshift_ai", "fix_state": "Not affected", "package_name": "rhoai/odh-llm-d-kv-cache-rhel9", "product_name": "Red Hat OpenShift AI (RHOAI)"}, {"cpe": "cpe:/a:redhat:openshift_ai", "fix_state": "Not affected", "package_name": "rhoai/odh-vllm-gaudi-rhel9", "product_name": "Red Hat OpenShift AI (RHOAI)"}], "public_date": "2026-07-06T19:41:11Z", "references": ["https://www.cve.org/CVERecord?id=CVE-2026-55646\nhttps://nvd.nist.gov/vuln/detail/CVE-2026-55646\nhttps://github.com/vllm-project/vllm/commit/b997071ec493765abbed990c65843ed05e4708a8\nhttps://github.com/vllm-project/vllm/pull/45510\nhttps://github.com/vllm-project/vllm/security/advisories/GHSA-v82g-2437-67m2"], "statement": "This Moderate denial of service flaw in vLLM, which could lead to excessive memory allocation when processing oversized audio files, does not affect Red Hat products. The vulnerable code is not present in Red Hat's supported offerings.", "threat_severity": "Moderate"}

{"affected": [{"configurations": [{"platform": null, "status": "affected", "versions": {"scheme": "semver", "value": "[0.22.0,0.24.0)"}}], "enrichment": {"confidence": 100.0, "confidence_source": "matching", "scores": [{"score": 100.0, "source": "matching"}]}, "original": {"product": "vllm", "source": "cna", "vendor": "vllm-project"}, "product": "vllm", "vendor": "vllm-project"}], "created": "2026-07-06T20:30:05.127282+00:00", "updated": "2026-08-01T18:30:04.996805+00:00", "vendors": ["vllm-project", "vllm-project$PRODUCT$vllm"]}