CVE-2026-111392
MEDIUM
6,3
Source: [email protected]
Attack Vector: network
Attack Complexity: high
Privileges Required: none
User Interaction: none
Confidentiality: N/A
Integrity: N/A
Availability: N/A
MEDIUM
4,8
Source: [email protected]
Attack Vector: network
Attack Complexity: high
Privileges Required: none
User Interaction: none
Scope: unchanged
Confidentiality: low
Integrity: low
Availability: none
Description
AI Translation Available
SGLang through 0.5.21 trusts caller-supplied mm_hashes values as the multimodal prefix-cache key, allowing remote attackers to poison cached KV entries across requests. Attackers can send a different image under a known or predicted hash via /generate to disclose another request's image content or corrupt victims' responses.
EPSS (Exploit Prediction Scoring System)
Trend Analysis
EPSS (Exploit Prediction Scoring System)
Prevede la probabilità di sfruttamento basata su intelligence sulle minacce e sulle caratteristiche della vulnerabilità.
EPSS Score
0,0021
Percentile
0,1th
Updated
EPSS Score Trend (Last 3 Days)
345
Insufficient Verification of Data Authenticity
DraftCommon Consequences
Security Scopes Affected:
Integrity
Other
Potential Impacts:
Varies By Context
Unexpected State
Applicable Platforms
Technologies:
ICS/OT
https://github.com/sgl-project/sglang/issues/44641
https://github.com/sgl-project/sglang/blob/v0.5.21/python/sglang/srt/managers/i…
https://github.com/sgl-project/sglang/blob/v0.5.21/python/sglang/srt/managers/t…
https://github.com/sgl-project/sglang/issues/44641
https://github.com/sgl-project/sglang/pull/44653
https://www.vulncheck.com/advisories/sglang-through-0.5.21-kv-cache-poisoning-v…