{"id":"CVE-2026-114055","aliases":["EUVD-2026-107006","GHSA-3cgq-c484-rx5j","GHSA-gx7p-2j49-hfq4"],"status":"published","title":"vLLM 0.7.3 before 0.31.0 DoS via Unvalidated mm_processor_kwargs","description":"vLLM from 0.7.3 before 0.31.0 contains an uncontrolled resource consumption vulnerability that allows remote clients to control multimodal processor parameters by supplying per-request mm_processor_kwargs. Attackers can set large image size, patch count, crop count, or frame count values in Isaac, InternVL, H2OVL, and other processors to exhaust memory and terminate the shared server.","severity":{"version":"4.0","score":8.7,"vector":"CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N","level":"high","assessor":"cna","from":"cve"},"epss":{"day":"2026-11-12","score":0.0042,"percentile":0.36778,"model":"v2026.06.15"},"facts":{"in_kev":false,"known_exploited":false,"public_exploit":false,"fix_available":true},"cwes":["CWE-400"],"published":"2026-10-21T00:36:31Z","modified":"2026-11-03T18:45:06Z","fixes":[{"ecosystem":"PyPI","package":"vllm","fixed_in":["0.31.0"],"sources":["ghsa","osv"]}],"affected":[{"source":"ghsa","ecosystem":"PyPI","package":"vllm","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0.7.3"},{"fixed":"0.31.0"}]}]},{"source":"osv","ecosystem":"PyPI","package":"vllm","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0.7.3"},{"fixed":"0.31.0"}]}],"versions":["0.10.0","0.10.1","0.10.1.1","0.10.2","0.11.0","0.11.1","0.11.2","0.12.0","0.13.0","0.14.0","0.14.1","0.15.0","0.15.1","0.16.0","0.17.0","0.17.1","0.18.0","0.18.1","0.19.0","0.19.1","0.20.0","0.20.1","0.20.2","0.21.0","0.22.0","0.22.1","0.23.0","0.24.0","0.25.0","0.25.1","0.26.0","0.27.0","0.27.1","0.28.0","0.29.0","0.30.0","0.7.3","0.8.0","0.8.1","0.8.2","0.8.3","0.8.4","0.8.5","0.8.5.post1","0.9.0","0.9.0.1","0.9.1","0.9.2"]}],"products":[{"vendor":"vllm","product":"vllm"},{"vendor":"vllm-project","product":"vllm"}],"references":[{"url":"https://github.com/vllm-project/vllm/security/advisories/GHSA-gx7p-2j49-hfq4","tags":["vendor-advisory","Vendor Advisory","advisory","web"]},{"url":"https://www.vulncheck.com/advisories/vllm-0.7.3-before-0.31.0-dos-via-unvalidated-mm-processor-kwargs","tags":["third-party-advisory","Third Party Advisory","advisory","web"]},{"url":"https://nvd.nist.gov/vuln/detail/CVE-2026-114055","tags":["advisory"]},{"url":"https://github.com/advisories/GHSA-3cgq-c484-rx5j"},{"url":"https://github.com/vllm-project/vllm/pull/58830","tags":["web"]},{"url":"https://github.com/vllm-project/vllm/commit/31f2e70cd3e403aaa471fb7fe3d33a7207aaf3cc","tags":["web"]},{"url":"https://github.com/vllm-project/vllm/releases/tag/v0.31.0","tags":["web"]},{"url":"https://github.com/advisories/GHSA-gx7p-2j49-hfq4"},{"url":"https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/114xxx/CVE-2026-114055.json","tags":["advisory"]},{"url":"https://github.com/vllm-project/vllm","tags":["package"]}],"sources":{"cve":[{"id":"CVE-2026-114055","title":"vLLM 0.7.3 before 0.31.0 DoS via Unvalidated mm_processor_kwargs","description":"vLLM from 0.7.3 before 0.31.0 contains an uncontrolled resource consumption vulnerability that allows remote clients to control multimodal processor parameters by supplying per-request mm_processor_kwargs. Attackers can set large image size, patch count, crop count, or frame count values in Isaac, InternVL, H2OVL, and other processors to exhaust memory and terminate the shared server.","status":"published","published":"2026-10-21T00:36:31.227Z","modified":"2026-10-26T19:49:33.622Z","cvss":[{"version":"4.0","score":8.7,"vector":"CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N","level":"high","assessor":"cna"},{"version":"3.1","score":7.5,"vector":"CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H","level":"high","assessor":"cna"}],"cwes":["CWE-400"],"updated_at":"2026-11-02T13:39:50.160666Z"}],"euvd":[{"id":"EUVD-2026-107006","description":"vLLM from 0.7.3 before 0.31.0 contains an uncontrolled resource consumption vulnerability that allows remote clients to control multimodal processor parameters by supplying per-request mm_processor_kwargs. Attackers can set large image size, patch count, crop count, or frame count values in Isaac, InternVL, H2OVL, and other processors to exhaust memory and terminate the shared server.","status":"published","published":"2026-10-21T00:36:31Z","modified":"2026-10-26T19:49:33Z","cvss":[{"version":"4.0","score":8.7,"vector":"CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N","level":"high"}],"updated_at":"2026-11-02T16:59:28.357127Z"}],"ghsa":[{"id":"GHSA-3cgq-c484-rx5j","title":"Duplicate Advisory: vLLM 0.7.3 before 0.31.0 DoS via Unvalidated mm_processor_kwargs","description":"## Duplicate Advisory\n\nThis advisory has been withdrawn because it is a duplicate of GHSA-gx7p-2j49-hfq4. This link is maintained to preserve external references.\n\n## Original Description\nvLLM from 0.7.3 before 0.31.0 contains an uncontrolled resource consumption vulnerability that allows remote clients to control multimodal processor parameters by supplying per-request mm_processor_kwargs. Attackers can set large image size, patch count, crop count, or frame count values in Isaac, InternVL, H2OVL, and other processors to exhaust memory and terminate the shared server.","status":"withdrawn","published":"2026-10-21T03:32:16Z","modified":"2026-10-23T20:58:04Z","cvss":[{"version":"3.1","score":7.5,"vector":"CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H","level":"high"},{"version":"4.0","score":8.7,"vector":"CVSS:4.0/AV:N/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","level":"high"}],"cwes":["CWE-400"],"updated_at":"2026-11-02T14:41:35.723785Z"},{"id":"GHSA-gx7p-2j49-hfq4","title":"vLLM : DoS via unvalidated mm_processor_kwargs in multimodal image processors (Isaac, InternVL, H2OVL, etc.)","description":"## Summary\n\nAPI clients can supply arbitrary `mm_processor_kwargs` per request. vLLM\nforwards these values through the multimodal rendering and processing pipeline\nto Hugging Face and custom image/video processor constructors and calls.\nResource-sensitive values such as image size, patch size, patch count, crop\ncount, or frame count can therefore be controlled by a remote client.\n\nFor processors that upscale inputs, a single request can force an allocation\nlarge enough to terminate the API server or worker with an out-of-memory\ncondition.\n\n## Affected Components\n\nThe per-request field was introduced in commit\n`041e29471671a3d43afedb6a749421d56ef4d3e7` and first shipped in v0.7.3.\nThe generic forwarding path remains present in current `main`.\n\nThe Isaac path is a concrete high-amplification example:\n\n- `ChatCompletionRequest.mm_processor_kwargs`\n- `OnlineRenderer` request extras\n- `MMProcessorInputs.hf_processor_mm_kwargs`\n- `BaseMultiModalProcessor.info.get_hf_processor(**hf_processor_mm_kwargs)`\n- `IsaacProcessingInfo.get_image_processor(**kwargs)`\n- `IsaacImageProcessor(**kwargs)`\n- `process_vision_for_patches()`\n\nEquivalent constructor forwarding exists in InternVL-derived processors,\nH2OVL, Step3VL, NemotronVL, and other multimodal processors. The issue is at\nthe shared remote-input boundary and is not limited to a fixed model list.\n\n## Impact\n\n- **Availability:** one request can exhaust CPU or GPU memory and terminate a\n  shared service process.\n- **Scope:** all users of the affected service instance lose availability.\n- **Authentication:** vLLM's API key is optional, so deployments without an\n  external authentication layer are remotely reachable without privileges.\n\n## Remediation\n\nReject non-empty per-request `mm_processor_kwargs` at every API-server request\nschema by default. Operators serving only trusted clients can explicitly opt\nin to the existing override behavior. Server-level `--mm-processor-kwargs` and\noffline `LLM` overrides remain available.\n\nThis shared boundary covers current and future multimodal processors instead\nof maintaining a model-specific list or a denylist that can miss custom\nprocessor arguments.\n\nA patch for this was merged here: https://github.com/vllm-project/vllm/pull/58830","status":"published","published":"2026-10-23T20:58:36Z","modified":"2026-10-23T20:58:38Z","cvss":[{"version":"3.1","score":7.5,"vector":"CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H","level":"high"}],"cwes":["CWE-400"],"updated_at":"2026-11-02T14:41:35.723785Z"}],"nvd":[{"id":"CVE-2026-114055","description":"vLLM from 0.7.3 before 0.31.0 contains an uncontrolled resource consumption vulnerability that allows remote clients to control multimodal processor parameters by supplying per-request mm_processor_kwargs. Attackers can set large image size, patch count, crop count, or frame count values in Isaac, InternVL, H2OVL, and other processors to exhaust memory and terminate the shared server.","status":"published","published":"2026-10-21T01:17:28.363Z","modified":"2026-11-03T18:45:06.357Z","cvss":[{"version":"4.0","score":8.7,"vector":"CVSS:4.0/AV:N/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","level":"high","assessor":"secondary"},{"version":"3.1","score":7.5,"vector":"CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H","level":"high","assessor":"secondary"}],"cwes":["CWE-400"],"updated_at":"2026-11-03T20:18:16.98942Z"}],"osv":[{"id":"CVE-2026-114055","title":"vLLM 0.7.3 before 0.31.0 DoS via Unvalidated mm_processor_kwargs","description":"vLLM from 0.7.3 before 0.31.0 contains an uncontrolled resource consumption vulnerability that allows remote clients to control multimodal processor parameters by supplying per-request mm_processor_kwargs. Attackers can set large image size, patch count, crop count, or frame count values in Isaac, InternVL, H2OVL, and other processors to exhaust memory and terminate the shared server.","status":"published","published":"2026-10-21T00:36:31.227Z","modified":"2026-11-05T03:29:48.93807504Z","cvss":[{"version":"4.0","score":8.7,"vector":"CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N","level":"high"}],"cwes":["CWE-400"],"updated_at":"2026-11-05T07:12:46.161235Z"},{"id":"GHSA-3cgq-c484-rx5j","title":"Duplicate Advisory: vLLM 0.7.3 before 0.31.0 DoS via Unvalidated mm_processor_kwargs","description":"## Duplicate Advisory\n\nThis advisory has been withdrawn because it is a duplicate of GHSA-gx7p-2j49-hfq4. This link is maintained to preserve external references.\n\n## Original Description\nvLLM from 0.7.3 before 0.31.0 contains an uncontrolled resource consumption vulnerability that allows remote clients to control multimodal processor parameters by supplying per-request mm_processor_kwargs. Attackers can set large image size, patch count, crop count, or frame count values in Isaac, InternVL, H2OVL, and other processors to exhaust memory and terminate the shared server.","status":"withdrawn","published":"2026-10-21T03:32:16Z","modified":"2026-10-23T21:15:03.768107947Z","cvss":[{"version":"3.1","score":7.5,"vector":"CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H","level":"high"},{"version":"4.0","score":8.7,"vector":"CVSS:4.0/AV:N/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","level":"high"}],"cwes":["CWE-400"],"updated_at":"2026-11-02T15:46:35.406455Z"},{"id":"GHSA-gx7p-2j49-hfq4","title":"vLLM : DoS via unvalidated mm_processor_kwargs in multimodal image processors (Isaac, InternVL, H2OVL, etc.)","description":"## Summary\n\nAPI clients can supply arbitrary `mm_processor_kwargs` per request. vLLM\nforwards these values through the multimodal rendering and processing pipeline\nto Hugging Face and custom image/video processor constructors and calls.\nResource-sensitive values such as image size, patch size, patch count, crop\ncount, or frame count can therefore be controlled by a remote client.\n\nFor processors that upscale inputs, a single request can force an allocation\nlarge enough to terminate the API server or worker with an out-of-memory\ncondition.\n\n## Affected Components\n\nThe per-request field was introduced in commit\n`041e29471671a3d43afedb6a749421d56ef4d3e7` and first shipped in v0.7.3.\nThe generic forwarding path remains present in current `main`.\n\nThe Isaac path is a concrete high-amplification example:\n\n- `ChatCompletionRequest.mm_processor_kwargs`\n- `OnlineRenderer` request extras\n- `MMProcessorInputs.hf_processor_mm_kwargs`\n- `BaseMultiModalProcessor.info.get_hf_processor(**hf_processor_mm_kwargs)`\n- `IsaacProcessingInfo.get_image_processor(**kwargs)`\n- `IsaacImageProcessor(**kwargs)`\n- `process_vision_for_patches()`\n\nEquivalent constructor forwarding exists in InternVL-derived processors,\nH2OVL, Step3VL, NemotronVL, and other multimodal processors. The issue is at\nthe shared remote-input boundary and is not limited to a fixed model list.\n\n## Impact\n\n- **Availability:** one request can exhaust CPU or GPU memory and terminate a\n  shared service process.\n- **Scope:** all users of the affected service instance lose availability.\n- **Authentication:** vLLM's API key is optional, so deployments without an\n  external authentication layer are remotely reachable without privileges.\n\n## Remediation\n\nReject non-empty per-request `mm_processor_kwargs` at every API-server request\nschema by default. Operators serving only trusted clients can explicitly opt\nin to the existing override behavior. Server-level `--mm-processor-kwargs` and\noffline `LLM` overrides remain available.\n\nThis shared boundary covers current and future multimodal processors instead\nof maintaining a model-specific list or a denylist that can miss custom\nprocessor arguments.\n\nA patch for this was merged here: https://github.com/vllm-project/vllm/pull/58830","status":"published","published":"2026-10-23T20:58:36Z","modified":"2026-10-23T21:15:03.776518731Z","cvss":[{"version":"3.1","score":7.5,"vector":"CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H","level":"high"}],"cwes":["CWE-400"],"updated_at":"2026-11-02T15:46:48.191583Z"}]}}