{"id":"CVE-2026-100653","aliases":["GHSA-hhv2-872h-628q","PYSEC-2026-4324"],"summary":"vLLM 0.22.1 before 0.28.0 Incomplete Artifact Pin Propagation","published_at":"2026-09-26T13:23:22.605Z","modified_at":"2026-10-23T19:00:39.873437Z","products":[],"rating":{"severity":"high","score":8.3,"vector":"CVSS:4.0/AV:N/AC:H/AT:P/PR:N/UI:N/VC:L/VI:H/VA:N/SC:N/SI:N/SA:N","attack_path":"internet_unauthenticated","user_interaction":false,"malicious":false,"red_hat":"Moderate","vendor_ratings":[{"source":"Red Hat","rating":"Moderate"}]},"fix_available":false,"kev":{"listed":false,"ransomware":false},"public_exploit":false,"exploits":[],"epss":{"score":0.0034000000450760126,"percentile":0.2799000144004822,"date":"2026-11-19T00:00:00Z"},"priority":10.3,"details":"vLLM is an inference and serving engine for large language models. In versions from 0.22.1 through 0.28.0, the operator-supplied model revision pin (--revision / --code-revision) is not propagated to several Hugging Face artifact loads for the FunAudioChat and Tarsier2 architectures: the WhisperFeatureExtractor and speech_tokenizer PreTrainedTokenizerFast loads in vllm/model_executor/models/funaudiochat.py and the Qwen2VLConfig.from_pretrained call used by Tarsier2ProcessingInfo in vllm/model_executor/models/qwen2_vl.py. As a result, deployments pinned to a reviewed revision still resolve these behavior-affecting processor, tokenizer, and config artifacts from the repository's default revision, so a later change to the upstream default branch can alter audio preprocessing, speech tokenizer behavior, or Tarsier2 configuration without any change to the operator's configured pin. This is a supply-chain integrity and reproducibility failure for pinned deployments; it is residual to the earlier fix tracked as GHSA-3ww4-5jv9-j5gm / CVE-2026-47155 and does not constitute remote code execution or a trust_remote_code=False bypass. The issue is fixed in version 0.28.0.","affected_stack":[],"affected_other":[{"ecosystem":"CVE","package":"vllm-project/vllm","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0.22.1"},{"fixed":"0.28.0"}]}],"fixed":["0.28.0"],"provider":"cvelistv5","source":"CVE-2026-100653","relation":"self"},{"ecosystem":"NVD","package":"vllm:vllm","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0.22.1"},{"fixed":"0.28.0"}]}],"fixed":["0.28.0"],"provider":"nvd","source":"CVE-2026-100653","relation":"self"},{"ecosystem":"PyPI","package":"vllm","purl":"pkg:pypi/vllm","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0.22.1"},{"fixed":"0.28.0"}]}],"versions":["0.22.1","0.23.0","0.24.0","0.25.0","0.25.1","0.26.0","0.27.0","0.27.1"],"fixed":["0.28.0"],"provider":"osv","source":"PYSEC-2026-4324","relation":"self"}],"product_status":[{"vendor":"Red Hat","product":"Red Hat AI Inference Server","cpe":"cpe:/a:redhat:ai_inference_server:3","status":"affected","packages":["rhaii/vllm-cpu-rhel9","rhaii/vllm-cuda-rhel9","rhaii/vllm-gaudi-rhel9","rhaii/vllm-neuron-rhel9","rhaii/vllm-rocm-rhel9","rhaii/vllm-spyre-rhel9","rhaii/vllm-tpu-rhel9","rhaiis/vllm-cpu-rhel9","rhaiis/vllm-cuda-rhel9","rhaiis/vllm-neuron-rhel9","rhaiis/vllm-rocm-rhel9","rhaiis/vllm-spyre-rhel9","rhaiis/vllm-tpu-rhel9"],"package_count":13,"fixed":[],"advisories":[],"note":"Fix deferred","provider":"redhat-vex","source":"CVE-2026-100653"},{"vendor":"Red Hat","product":"Red Hat Enterprise Linux AI (RHEL AI) 3","cpe":"cpe:/a:redhat:enterprise_linux_ai:3","status":"affected","packages":["rhelai3/bootc-aws-cuda-rhel9","rhelai3/bootc-azure-cuda-rhel9","rhelai3/bootc-azure-rocm-rhel9","rhelai3/bootc-cuda-rhel9","rhelai3/bootc-gaudi-rhel9","rhelai3/bootc-gcp-cuda-rhel9","rhelai3/bootc-rocm-rhel9"],"package_count":7,"fixed":[],"advisories":[],"note":"Fix deferred","provider":"redhat-vex","source":"CVE-2026-100653"},{"vendor":"Red Hat","product":"Red Hat OpenShift AI (RHOAI)","cpe":"cpe:/a:redhat:openshift_ai","status":"affected","packages":["rhoai/odh-kserve-agent-rhel9","rhoai/odh-kserve-controller-rhel9","rhoai/odh-kserve-router-rhel9","rhoai/odh-kserve-storage-initializer-rhel9","rhoai/odh-llm-d-kv-cache-rhel9","rhoai/odh-th-torch-cuda-py312-rhel9","rhoai/odh-vllm-gaudi-rhel9"],"package_count":7,"fixed":[],"advisories":[],"note":"Fix deferred","provider":"redhat-vex","source":"CVE-2026-100653"}],"records":[{"provider":"csaf-bsi","source_id":"WID-SEC-W-2026-3327","relation":"upstream","url":"https://wid.cert-bund.de/portal/wid/securityadvisory?name=WID-SEC-2026-3327","license":"TLP:CLEAR, mirroring permitted","published_at":"2026-09-13T22:00:00Z","modified_at":"2026-11-09T23:00:00Z","summary":"vllm: Mehrere Schwachstellen ermöglichen","details":"Ein Angreifer kann mehrere Schwachstellen in vllm ausnutzen, um Sicherheitsvorkehrungen zu umgehen, Daten zu manipulieren, sensible Informationen offenzulegen oder einen Denial-of-Service-Zustand auszulösen.\n\nOpen Source vLLM ist eine Open-Source-Bibliothek für schnelle und effiziente Inferenz von Large Language Models (LLMs).","severity":[{"type":"BSI","score":"mittel","source":"WID-SEC-W-2026-3327"}],"affected":[],"product_status":[],"references":[{"type":"WEB","url":"https://wid.cert-bund.de/.well-known/csaf/white/2026/wid-sec-w-2026-3327.json"},{"type":"WEB","url":"https://wid.cert-bund.de/portal/wid/securityadvisory?name=WID-SEC-2026-3327"},{"type":"WEB","url":"https://github.com/vllm-project/vllm/security/advisories/GHSA-3mqx-f33v-vgp9"},{"type":"WEB","url":"https://github.com/vllm-project/vllm/security/advisories/GHSA-hhv2-872h-628q"},{"type":"WEB","url":"https://github.com/vllm-project/vllm/security/advisories/GHSA-j682-9xp5-rrf3"},{"type":"WEB","url":"https://github.com/vllm-project/vllm/security/advisories/GHSA-jcq2-4gch-5qhf"},{"type":"WEB","url":"https://github.com/vllm-project/vllm/security/advisories/GHSA-p6g9-7v3x-m8mv"},{"type":"WEB","url":"https://github.com/vllm-project/vllm/security/advisories/GHSA-qff2-492f-9fm4"},{"type":"WEB","url":"https://github.com/vllm-project/vllm/security/advisories/GHSA-v5gm-qgmv-gc6c"},{"type":"WEB","url":"https://github.com/vllm-project/vllm/security/advisories/GHSA-wpww-v874-ph2p"},{"type":"WEB","url":"https://access.redhat.com/errata/RHSA-2026:90440"},{"type":"WEB","url":"https://access.redhat.com/errata/RHSA-2026:94721"},{"type":"WEB","url":"https://access.redhat.com/errata/RHSA-2026:94702"},{"type":"WEB","url":"https://access.redhat.com/errata/RHSA-2026:94709"},{"type":"WEB","url":"https://access.redhat.com/errata/RHSA-2026:94738"},{"type":"WEB","url":"https://access.redhat.com/errata/RHSA-2026:94717"},{"type":"WEB","url":"https://access.redhat.com/errata/RHSA-2026:96323"},{"type":"WEB","url":"https://access.redhat.com/errata/RHSA-2026:96333"}]},{"provider":"cvelistv5","source_id":"CVE-2026-100653","relation":"self","url":"https://www.cve.org/CVERecord?id=CVE-2026-100653","license":"CC-BY-4.0","published_at":"2026-09-26T13:23:22.605Z","modified_at":"2026-10-01T11:06:45.139Z","summary":"vLLM 0.22.1 before 0.28.0 Incomplete Artifact Pin Propagation","details":"vLLM is an inference and serving engine for large language models. In versions from 0.22.1 through 0.28.0, the operator-supplied model revision pin (--revision / --code-revision) is not propagated to several Hugging Face artifact loads for the FunAudioChat and Tarsier2 architectures: the WhisperFeatureExtractor and speech_tokenizer PreTrainedTokenizerFast loads in vllm/model_executor/models/funaudiochat.py and the Qwen2VLConfig.from_pretrained call used by Tarsier2ProcessingInfo in vllm/model_executor/models/qwen2_vl.py. As a result, deployments pinned to a reviewed revision still resolve these behavior-affecting processor, tokenizer, and config artifacts from the repository's default revision, so a later change to the upstream default branch can alter audio preprocessing, speech tokenizer behavior, or Tarsier2 configuration without any change to the operator's configured pin. This is a supply-chain integrity and reproducibility failure for pinned deployments; it is residual to the earlier fix tracked as GHSA-3ww4-5jv9-j5gm / CVE-2026-47155 and does not constitute remote code execution or a trust_remote_code=False bypass. The issue is fixed in version 0.28.0.","severity":[{"type":"CVSS_V4","score":"CVSS:4.0/AV:N/AC:H/AT:P/PR:N/UI:N/VC:L/VI:H/VA:N/SC:N/SI:N/SA:N","source":"CVE-2026-100653"},{"type":"CVSS_V3","score":"CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:L/I:H/A:N","source":"CVE-2026-100653"}],"affected":[{"ecosystem":"CVE","package":"vllm-project/vllm","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0.22.1"},{"fixed":"0.28.0"}]}],"fixed":["0.28.0"],"provider":"cvelistv5","source":"CVE-2026-100653","relation":"self"}],"product_status":[],"references":[{"type":"ADVISORY","url":"https://github.com/vllm-project/vllm/security/advisories/GHSA-hhv2-872h-628q"},{"type":"FIX","url":"https://github.com/vllm-project/vllm/commit/d26a28ab033697f55a1414b5b0435de7cd6045b6"},{"type":"WEB","url":"https://www.vulncheck.com/advisories/vllm-0.22.1-before-0.28.0-incomplete-artifact-pin-propagation"}]},{"provider":"nvd","source_id":"CVE-2026-100653","relation":"self","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-100653","severity":[],"affected":[{"ecosystem":"NVD","package":"vllm:vllm","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0.22.1"},{"fixed":"0.28.0"}]}],"fixed":["0.28.0"],"provider":"nvd","source":"CVE-2026-100653","relation":"self"}],"product_status":[],"references":[]},{"provider":"osv","source_id":"PYSEC-2026-4324","relation":"self","url":"https://osv.dev/vulnerability/PYSEC-2026-4324","license":"CC-BY-4.0","published_at":"2026-09-26T14:16:47.953Z","modified_at":"2026-10-23T19:00:39.873437Z","details":"vLLM is an inference and serving engine for large language models. In versions from 0.22.1 through 0.28.0, the operator-supplied model revision pin (--revision / --code-revision) is not propagated to several Hugging Face artifact loads for the FunAudioChat and Tarsier2 architectures: the WhisperFeatureExtractor and speech_tokenizer PreTrainedTokenizerFast loads in vllm/model_executor/models/funaudiochat.py and the Qwen2VLConfig.from_pretrained call used by Tarsier2ProcessingInfo in vllm/model_executor/models/qwen2_vl.py. As a result, deployments pinned to a reviewed revision still resolve these behavior-affecting processor, tokenizer, and config artifacts from the repository's default revision, so a later change to the upstream default branch can alter audio preprocessing, speech tokenizer behavior, or Tarsier2 configuration without any change to the operator's configured pin. This is a supply-chain integrity and reproducibility failure for pinned deployments; it is residual to the earlier fix tracked as GHSA-3ww4-5jv9-j5gm / CVE-2026-47155 and does not constitute remote code execution or a trust_remote_code=False bypass. The issue is fixed in version 0.28.0.","severity":[{"type":"CVSS_V4","score":"CVSS:4.0/AV:N/AC:H/AT:P/PR:N/UI:N/VC:L/VI:H/VA:N/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","source":"PYSEC-2026-4324"}],"affected":[{"ecosystem":"PyPI","package":"vllm","purl":"pkg:pypi/vllm","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0.22.1"},{"fixed":"0.28.0"}]}],"versions":["0.22.1","0.23.0","0.24.0","0.25.0","0.25.1","0.26.0","0.27.0","0.27.1"],"fixed":["0.28.0"],"provider":"osv","source":"PYSEC-2026-4324","relation":"self"}],"product_status":[],"references":[{"type":"ADVISORY","url":"https://www.vulncheck.com/advisories/vllm-0.22.1-before-0.28.0-incomplete-artifact-pin-propagation"},{"type":"FIX","url":"https://github.com/vllm-project/vllm/commit/d26a28ab033697f55a1414b5b0435de7cd6045b6"},{"type":"EVIDENCE","url":"https://github.com/vllm-project/vllm/security/advisories/GHSA-hhv2-872h-628q"}]},{"provider":"redhat-vex","source_id":"CVE-2026-100653","relation":"self","url":"https://access.redhat.com/security/cve/CVE-2026-100653","license":"CC-BY-4.0","published_at":"2026-09-26T13:23:22.605Z","modified_at":"2026-10-01T16:40:16Z","summary":"vllm: vLLM: Supply chain integrity compromise via incomplete model revision pinning","details":"A flaw was found in vLLM. When operators configure a deployment to use a specific, pinned model revision, vLLM fails to apply the revision pin to certain supporting artifacts, such as model processors and configuration files. Consequently, these components are loaded from the external repository's default branch rather than the approved revision. A remote attacker with control over the upstream repository could exploit this flaw to alter model behavior and compromise deployment integrity without the operator's knowledge.","severity":[{"type":"CVSS_V3","score":"CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:L/I:H/A:N","source":"CVE-2026-100653"},{"type":"Red Hat","score":"Moderate","source":"CVE-2026-100653"}],"affected":[],"product_status":[{"vendor":"Red Hat","product":"Red Hat AI Inference Server","cpe":"cpe:/a:redhat:ai_inference_server:3","status":"affected","packages":["rhaii/vllm-cpu-rhel9","rhaii/vllm-cuda-rhel9","rhaii/vllm-gaudi-rhel9","rhaii/vllm-neuron-rhel9","rhaii/vllm-rocm-rhel9","rhaii/vllm-spyre-rhel9","rhaii/vllm-tpu-rhel9","rhaiis/vllm-cpu-rhel9","rhaiis/vllm-cuda-rhel9","rhaiis/vllm-neuron-rhel9","rhaiis/vllm-rocm-rhel9","rhaiis/vllm-spyre-rhel9","rhaiis/vllm-tpu-rhel9"],"package_count":13,"fixed":[],"advisories":[],"note":"Fix deferred","provider":"redhat-vex","source":"CVE-2026-100653"},{"vendor":"Red Hat","product":"Red Hat Enterprise Linux AI (RHEL AI) 3","cpe":"cpe:/a:redhat:enterprise_linux_ai:3","status":"affected","packages":["rhelai3/bootc-aws-cuda-rhel9","rhelai3/bootc-azure-cuda-rhel9","rhelai3/bootc-azure-rocm-rhel9","rhelai3/bootc-cuda-rhel9","rhelai3/bootc-gaudi-rhel9","rhelai3/bootc-gcp-cuda-rhel9","rhelai3/bootc-rocm-rhel9"],"package_count":7,"fixed":[],"advisories":[],"note":"Fix deferred","provider":"redhat-vex","source":"CVE-2026-100653"},{"vendor":"Red Hat","product":"Red Hat OpenShift AI (RHOAI)","cpe":"cpe:/a:redhat:openshift_ai","status":"affected","packages":["rhoai/odh-kserve-agent-rhel9","rhoai/odh-kserve-controller-rhel9","rhoai/odh-kserve-router-rhel9","rhoai/odh-kserve-storage-initializer-rhel9","rhoai/odh-llm-d-kv-cache-rhel9","rhoai/odh-th-torch-cuda-py312-rhel9","rhoai/odh-vllm-gaudi-rhel9"],"package_count":7,"fixed":[],"advisories":[],"note":"Fix deferred","provider":"redhat-vex","source":"CVE-2026-100653"}],"references":[{"type":"WEB","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100653.json"},{"type":"WEB","url":"https://access.redhat.com/security/cve/CVE-2026-100653"},{"type":"WEB","url":"https://bugzilla.redhat.com/show_bug.cgi?id=2541818"},{"type":"WEB","url":"https://www.cve.org/CVERecord?id=CVE-2026-100653"},{"type":"WEB","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-100653"},{"type":"WEB","url":"https://github.com/vllm-project/vllm/commit/d26a28ab033697f55a1414b5b0435de7cd6045b6"},{"type":"WEB","url":"https://github.com/vllm-project/vllm/security/advisories/GHSA-hhv2-872h-628q"},{"type":"WEB","url":"https://www.vulncheck.com/advisories/vllm-0.22.1-before-0.28.0-incomplete-artifact-pin-propagation"}]}],"references":[{"type":"ADVISORY","url":"https://github.com/vllm-project/vllm/security/advisories/GHSA-hhv2-872h-628q"},{"type":"FIX","url":"https://github.com/vllm-project/vllm/commit/d26a28ab033697f55a1414b5b0435de7cd6045b6"},{"type":"WEB","url":"https://www.vulncheck.com/advisories/vllm-0.22.1-before-0.28.0-incomplete-artifact-pin-propagation"},{"type":"WEB","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100653.json"},{"type":"WEB","url":"https://access.redhat.com/security/cve/CVE-2026-100653"},{"type":"WEB","url":"https://bugzilla.redhat.com/show_bug.cgi?id=2541818"},{"type":"WEB","url":"https://www.cve.org/CVERecord?id=CVE-2026-100653"},{"type":"WEB","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-100653"}],"sources":[{"provider":"csaf-bsi","source_id":"WID-SEC-W-2026-3327","url":"https://wid.cert-bund.de/portal/wid/securityadvisory?name=WID-SEC-2026-3327","license":"TLP:CLEAR, mirroring permitted","modified_at":"2026-11-09T23:00:00Z","relation":"upstream"},{"provider":"cvelistv5","source_id":"CVE-2026-100653","url":"https://www.cve.org/CVERecord?id=CVE-2026-100653","license":"CC-BY-4.0","modified_at":"2026-10-01T11:06:45.139Z","relation":"self"},{"provider":"nvd","source_id":"CVE-2026-100653","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-100653","relation":"self"},{"provider":"osv","source_id":"PYSEC-2026-4324","url":"https://osv.dev/vulnerability/PYSEC-2026-4324","license":"CC-BY-4.0","modified_at":"2026-10-23T19:00:39.873437Z","relation":"self"},{"provider":"redhat-vex","source_id":"CVE-2026-100653","url":"https://access.redhat.com/security/cve/CVE-2026-100653","license":"CC-BY-4.0","modified_at":"2026-10-01T16:40:16Z","relation":"self"}]}
