{"id":"CVE-2026-100650","state":"PUBLISHED","title":"vLLM before 0.29.0 Resource Exhaustion via Unbounded Media Materialization","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.","publishedAt":"2026-09-26T13:23:20.418Z","updatedAt":"2026-10-01T11:06:44.741Z","severity":{"score":7.1,"label":"High","version":"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","source":"VulnCheck"},"assigner":"VulnCheck","products":["vllm-project vllm"],"affected":[{"vendor":"vllm-project","product":"vllm","packageURL":"pkg:pypi/vllm","defaultStatus":"unaffected","versions":[{"version":"0","status":"affected","lessThan":"0.29.0","versionType":"semver"},{"version":"0.29.0","status":"unaffected","versionType":"semver"}]}],"references":["https://github.com/vllm-project/vllm/security/advisories/GHSA-p6g9-7v3x-m8mv","https://github.com/vllm-project/vllm/commit/752a3a504485790a2e8491cacbb35c137339ad34","https://www.vulncheck.com/advisories/vllm-before-0.29.0-resource-exhaustion-via-unbounded-media-materialization"],"weaknesses":["CWE-400"],"ingestedAt":"2026-10-29T20:10:13.529Z","source":{"repository":"CVEProject/cvelistV5","path":"cves/2026/100xxx/CVE-2026-100650.json","commit":"556c9d3a8cd17e218bb8e452ba7cd619a384ff47"},"raw":{"dataType":"CVE_RECORD","dataVersion":"5.2","cveMetadata":{"cveId":"CVE-2026-100650","state":"PUBLISHED","datePublished":"2026-09-26T13:23:20.418Z","dateUpdated":"2026-10-01T11:06:44.741Z","assignerShortName":"VulnCheck","assignerOrgId":"83251b91-4cc7-4094-a5c7-464a1b83ea10","dateReserved":"2026-09-26T02:33:07.898Z"},"containers":{"cna":{"title":"vLLM before 0.29.0 Resource Exhaustion via Unbounded Media Materialization","descriptions":[{"lang":"en","value":"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."}],"affected":[{"vendor":"vllm-project","product":"vllm","packageURL":"pkg:pypi/vllm","defaultStatus":"unaffected","versions":[{"version":"0","status":"affected","lessThan":"0.29.0","versionType":"semver"},{"version":"0.29.0","status":"unaffected","versionType":"semver"}]}],"references":[{"url":"https://github.com/vllm-project/vllm/security/advisories/GHSA-p6g9-7v3x-m8mv","tags":["vendor-advisory"],"name":"GitHub Security Advisory (GHSA-p6g9-7v3x-m8mv)"},{"url":"https://github.com/vllm-project/vllm/commit/752a3a504485790a2e8491cacbb35c137339ad34","tags":["patch"],"name":"Patch Commit"},{"url":"https://www.vulncheck.com/advisories/vllm-before-0.29.0-resource-exhaustion-via-unbounded-media-materialization","name":"VulnCheck Advisory: vLLM before 0.29.0 Resource Exhaustion via Unbounded Media Materialization","tags":["third-party-advisory"]}],"metrics":[{"format":"CVSS","cvssV4_0":{"version":"4.0","vectorString":"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","attackVector":"NETWORK","attackComplexity":"LOW","attackRequirements":"NONE","privilegesRequired":"LOW","userInteraction":"NONE","vulnConfidentialityImpact":"NONE","vulnIntegrityImpact":"NONE","vulnAvailabilityImpact":"HIGH","subConfidentialityImpact":"NONE","subIntegrityImpact":"NONE","subAvailabilityImpact":"NONE","baseScore":7.1,"baseSeverity":"HIGH"}},{"format":"CVSS","cvssV3_1":{"version":"3.1","vectorString":"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H","attackVector":"NETWORK","attackComplexity":"LOW","privilegesRequired":"LOW","userInteraction":"NONE","scope":"UNCHANGED","confidentialityImpact":"NONE","integrityImpact":"NONE","availabilityImpact":"HIGH","baseScore":6.5,"baseSeverity":"MEDIUM"}}],"problemTypes":[{"descriptions":[{"cweId":"CWE-400","lang":"en","description":"Uncontrolled Resource Consumption","type":"CWE"}]}],"providerMetadata":{"shortName":"VulnCheck","orgId":"83251b91-4cc7-4094-a5c7-464a1b83ea10","dateUpdated":"2026-10-01T11:06:44.741Z"},"datePublic":"2026-09-12T00:00:00.000Z","cpeApplicability":[{"nodes":[{"operator":"OR","negate":false,"cpeMatch":[{"vulnerable":true,"criteria":"cpe:2.3:a:vllm:vllm:*:*:*:*:*:*:*:*","versionEndExcluding":"0.29.0"}]}]}],"credits":[{"lang":"en","value":"KernelClint","type":"finder"},{"lang":"en","value":"dhalf","type":"finder"},{"lang":"en","value":"jperezdealgaba","type":"coordinator"}],"x_generator":{"engine":"vulncheck-endgame"}},"adp":[{"title":"CISA ADP Vulnrichment","references":[{"url":"https://github.com/vllm-project/vllm/security/advisories/GHSA-p6g9-7v3x-m8mv","tags":["exploit"]}],"metrics":[{"other":{"type":"ssvc","content":{"timestamp":"2026-09-30T14:49:40.322864Z","id":"CVE-2026-100650","options":[{"Exploitation":"poc"},{"Automatable":"no"},{"Technical Impact":"partial"}],"role":"CISA Coordinator","version":"2.0.3"}}}],"providerMetadata":{"shortName":"CISA-ADP","orgId":"134c704f-9b21-4f2e-91b3-4a467353bcc0","dateUpdated":"2026-09-30T14:50:14.406Z"}}]}}}