vLLM: Mirrored multimodal IPC caches desync after a rejected request — a later request reusing the same media hash trips a receiver assertion in the engine core

[ Description ]Official description

vLLM is an inference and serving engine for large language models. Prior to 0.28.0, the default mirrored multimodal LRU cache can commit a media hash in the frontend sender cache during multimodal rendering and before engine admission, while the engine receiver cache never receives the payload if that request is rejected. A later request reusing the same media hash causes MultiModalProcessorSenderCache to send no payload and MultiModalReceiverCache to reach an assertion with the message "Expected a cached item," producing a shared-service availability failure. This issue is fixed in version 0.28.0.

[ Fixes ]Fixes and patches

• vllm 0.28.0Fixed version · GitHub Advisories · pip package vllm: upgrade to 0.28.0 or later (affected < 0.28.0)
• Patch on github.comVendor patch or bulletin · NVD · Reference tagged 'Patch' by NVD
• Patch on github.comVendor patch or bulletin · NVD · Reference tagged 'Patch' by NVD
• Patch on github.comVendor patch or bulletin · NVD · Reference tagged 'Patch' by NVD

[ Advisories ]Government CERT advisories

• BSI CERT-Bund5 Oct 2026WID-SEC-2026-3754

  vllm: Multiple vulnerabilities allow denial of service Translated from German

  CVE-2026-105758CVE-2026-105760CVE-2026-105759+2 more CVEs

[ Packages ]Affected open-source packages

• vllm pipAffected < 0.28.0 · fixed in 0.28.0 · GHSA-ph3r-5jfg-f84f

[ References ]References

• github.com/vllm-project/vllm/commit/396204230423b7cc6798300926b8fa30190d26a9Patch
• github.com/vllm-project/vllm/pull/46747Issue TrackingPatch
• github.com/vllm-project/vllm/pull/51897Issue TrackingPatch
• github.com/vllm-project/vllm/releases/tag/v0.28.0Release Notes
• github.com/vllm-project/vllm/security/advisories/GHSA-ph3r-5jfg-f84fMitigationVendor Advisory