vLLM: Scale-out disaggregated multimodal transport trusts caller-supplied features
CVE-2026-105754 · GHSA-ph72-cqr5-qpp7 · published 2026-10-05
- Severity
- Medium
- CVSS score
- 6.5
- Chance of exploitation (EPSS)
- 0.27%
- Fixed in
- 0.30.0
- Exploitation No known exploitation
- Automatable No
- Technical impact Partial
- CISA SSVC
A single authenticated request to a scale-out multimodal deployment can: Crash the shared EngineCore process (sites 1, 2, 3), taking the served model down for every tenant (/health → 503). Availability-only; no code execution or data disclosure demonstrated for these sites. Silently poison or read back another request's shared encoder-cache state (site 4) — an integrity/disclosure primitive. Attack complexity is High because the attacker must know or induce the victim's content hash; there is no availability impact for this site. Corrupt backend-visible placeholder semantics across the render-to-generate boundary (site 5) for models that depend on sparse is_embed masks.
Upgrade vllm to 0.30.0 or later. If you start it from an MCP config with npx or uvx, pin that version so an old cached copy is not used.
On the scale-out path, do not trust caller-supplied multimodal state as renderer-produced. After decoding features, rebind and validate the reconstructed MultiModalKwargsItem against the active model's renderer contract at the HTTP boundary: 1. Reject any request whose decoded grid geometry is inconsistent with the rendered pixel-tensor element count and declared placeholder span, before it reaches prepareencodermetadata() (site 1). 2. Rebind each field's processor type to the schema the active model's renderer declares (or reject if it differs), instead of reconstructing internal field processors from wire-selected factory names (site 2). 3.
| Package | Registry | Affected | Fixed in |
|---|---|---|---|
| vllm | PyPI | < 0.30.0 | 0.30.0 |
- CriticalMarimo: Pre-Auth Remote Code Execution via Terminal WebSocket Authentication BypassmarimoExploited
- CriticalLangflow CORS misconfiguration enables Account Takeover and RCElangflowExploited
- HighLiteLLM: MCP Authentication Bypass via OAuth2 Passthrough FallbacklitellmExploited
- HighLangflow: IDOR Vulnerability in `/api/v1/responses` Endpoint Allows Authenticated Attackers to Access Another User's FlowlangflowExploited
- CriticalTodoist MCP Server: Browser DNS rebinding lets websites write and delete Todoist data through local MCP server@doist/todoist-mcp
- CriticalLangflow: Unauthenticated Flow Execution via Webhook Authentication Bypasslangflow
- CriticalvLLM: OpenAI auth bypassvllm
- CriticalvLLM has RCE In Video Processingvllm
- CriticalvLLM Allows Remote Code Execution via PyNcclPipe Communication Servicevllm
- CriticalvLLM Vulnerable to Remote Code Execution via Mooncake Integrationvllm
- CriticalCVE-2025-24357 Malicious model remote code execution fix bypass with PyTorch < 2.6.0vllm
- CriticalvLLM Deserialization of Untrusted Data vulnerabilityvllm
The impact, patch and workaround text comes from the advisory, shortened (GitHub Advisory Database, CC BY 4.0). Proof-of-concept and reproduction sections and code are left out; the original advisory has them.