EUVD-2026-87747
GHSA-r3cf-2wgr-7xh9
Severity by source
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
Network-reachable but AC:H due to required upstream-repo control and specific vulnerable architectures; integrity-only (I:H) as pinned artifacts are silently substituted, no confidentiality or availability impact.
Primary rating from Vendor (VulnCheck).
CVSS VectorVendor: VulnCheck
Lifecycle Timeline
5DescriptionCVE.org
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.
Articles & Coverage 1
AnalysisAI
Pinned vLLM deployments (0.22.1 through 0.28.0) running the FunAudioChat or Tarsier2 architectures silently ignore the operator's --revision/--code-revision pin for several Hugging Face artifacts, allowing a subsequent change to the upstream repository's default branch to alter audio preprocessing, speech-tokenizer behavior, or Tarsier2 configuration without any change to the operator's configured pin. The result is a supply-chain integrity and reproducibility failure rather than a code-execution flaw: the advisory explicitly states it is residual to CVE-2026-47155 / GHSA-3ww4-5jv9-j5gm and does not constitute remote code execution or a trust_remote_code=False bypass, despite an 'RCE' tag appearing in some threat-feed metadata. …
Unlock full vulnerability intelligence
- Risk assessment & exploitation conditions
- Attack chain visualization
- Remediation with exact patch versions
- Threat intelligence from 22 sources
- Personal watchlist & email alerts
No credit card · 7-day full trial
Attack ChainAIDerived
Hypothetical attack flow derived from CVE metadata
Vulnerability AssessmentAI
| Exploitation | Requires ALL of: (1) deployment running the FunAudioChat or Tarsier2 model architectures specifically - no other architectures are affected by this residual gap; (2) the operator using a --revision or --code-revision pin (the flaw is that the pin is silently NOT honored for the WhisperFeatureExtractor, speech_tokenizer, and Qwen2VLConfig.from_pretrained loads); and (3) an actor able to change the upstream Hugging Face repository's DEFAULT branch/revision, since those artifacts resolve from the default rather than the pin. … Additional conditions and limiting factors are described in the full assessment. |
| Risk Assessment | This is a genuine but low-urgency issue: a supply-chain integrity/reproducibility gap, NOT a remote code execution flaw despite the misleading 'RCE' input tag (the advisory explicitly states it 'does not constitute remote code execution or a trust_remote_code=False bypass'). … Full risk analysis with EPSS, KEV, and SSVC signal comparison available after sign-in. |
| Exploit Scenario | Full exploit scenario with step-by-step reproduction available after sign-in. |
| Remediation | Vendor-released patch: upgrade to vLLM 0.28.0, which propagates model_config.revision and model_config.code_revision into the affected artifact loads (fixed by upstream commit d26a28ab033697f55a1414b5b0435de7cd6045b6, referenced from the advisory at https://github.com/vllm-project/vllm/security/advisories/GHSA-hhv2-872h-628q). … Detailed patch versions, workarounds, and compensating controls in full report. |
Recommended ActionAI
Within 24 hours, inventory every vLLM deployment and identify instances running versions 0.22.1 through 0.28.0 that serve the FunAudioChat or Tarsier2 architectures, confirm whether --revision or --code-revision pins are in use on those workloads, and record the current artifact identifiers and content hashes for each model pulled from Hugging Face so a later drift can be proven; also confirm whether the remediation for CVE-2026-47155 / GHSA-3ww4-5jv9-j5gm was fully applied, since this issue is residual to it. …
Sign in for detailed remediation steps and compensating controls.
Threat intelligence, references, and detailed analysis are available after sign-in.
Wazuh SIEM platform versions 4.4.0 through 4.9.0 contain an unsafe deserialization vulnerability in the DistributedAPI t
BentoML version 1.4.2 and earlier contains an unauthenticated remote code execution vulnerability through insecure deser
pgAdmin 4 contains critical remote code execution vulnerabilities in the Query Tool download and Cloud Deployment endpoi
The renderLocalView function in render/views.py in graphite-web in Graphite 0.9.5 through 0.9.10 uses the pickle Python
BentoML is a Python library for building online serving systems optimized for AI apps and model inference. Rated critica
OpenSSL before 0.9.8za, 1.0.0 before 1.0.0m, and 1.0.1 before 1.0.1h does not properly restrict processing of ChangeCiph
pyLoad download manager version prior to 0.5.0b3.dev77 exposes the Flask SECRET_KEY through an unauthenticated endpoint.
Langflow (a visual LLM pipeline builder) contains a critical unauthenticated code execution vulnerability (CVE-2026-3301
In Mercurial before 4.1.3, "hg serve --stdio" allows remote authenticated users to launch the Python debugger, and conse
Unauthenticated remote code execution affects Kestra OSS (the open-source event-driven orchestration platform) prior to
Unauthenticated remote code execution in Marimo ≤0.20.4 allows attackers to execute arbitrary system commands via the `/
pyLoad is the free and open-source Download Manager written in pure Python. Rated medium severity (CVSS 5.3), this vulne
Same weakness CWE-348 – Use of Less Trusted Source
View allVendor StatusVendor
Share
External POC / Exploit Code
Leaving vuln.today