CVE-2026-105752

Description

vLLM is an inference and serving engine for large language models. Prior to 0.30.0, Harmony tool continuations submitted through "POST /v1/responses" requests rebuild the next-turn engine input without preserving the cache_salt value, placing the continuation prefix in the global unsalted cache namespace even when the caller enabled salting. On deployments with prefix caching enabled, which is the default, an authenticated tenant who can reconstruct a victim's low-entropy post-tool history can submit the same continuation and use the cached_tokens_per_turn count to determine whether the prefix was previously processed, defeating the intended tenant isolation of salted prefix caching. This issue is fixed in version 0.30.0.

Traduire en français

Produits concernés

• vllm-project vllm
• Vllm vllm

Scores CVSS

Faiblesses (CWE)

• CWE-200
• CWE-524

Références

• https://github.com/vllm-project/vllm/commit/6a2a2bb02b563b83f946012959fd3927984d072a Patch
• https://github.com/vllm-project/vllm/pull/50195 Issue TrackingPatch
• https://github.com/vllm-project/vllm/pull/51818 Issue TrackingPatch
• https://github.com/vllm-project/vllm/releases/tag/v0.30.0 Release Notes
• https://github.com/vllm-project/vllm/security/advisories/GHSA-935w-9g4m-p28p MitigationVendor Advisory

Ailleurs

• NVD
• CVE.org
• CIRCL Vulnerability-Lookup
• CERT-FR
• GitHub Advisories