CVE-2026-105752Low· 3.1▾ SunlitvLLM 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, pl…
▾ Sunlit zone — Low / medium · no exploitation signal
impact 17.1 · likelihood 0 · exploitation 0
Need a working PoC? Pro members can cast a request and our team develops one — it lands right here.
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.
Refer to the linked advisories for vendor-supplied fixes and affected version ranges.
Affected packages:
vllm < 0.30.0Patched in:
vllm 0.30.0Connected by shared product, vendor, weakness, or advisory.
CVE-2026-105760Medium· 5.3vLLM is an inference and serving engine for large language models
CVE-2026-105758Medium· 5.3vLLM is an inference and serving engine for large language models
CVE-2026-105757Medium· 6.5vLLM is an inference and serving engine for large language models
CVE-2026-105754Medium· 6.5vLLM is an inference and serving engine for large language models
CVE-2026-105755Medium· 4.2vLLM is an inference and serving engine for large language models
CVE-2026-105756Medium· 6.5vLLM is an inference and serving engine for large language models