Answer in brief
CVE-2026-111392 records a Medium severity (CVSS 6.3) vulnerability in SGLang through 0.5.21 KV Cache Poisoning via mm_hashes in /generate. The current sources do not mark it as known exploited. The current feed maps sgl-project/sglang (generic). Check affected ranges and fixed versions before updating.
Analysis pending evidence review
HOL Guard separates source facts from reviewed analysis. See the methodology.
CVSS is 6.3. The current sources do not mark it as known exploited. Treat this as a source-backed prioritization signal, not a statement about your environment.
Analysis status
Analysis pending evidence review
Factual feed record only; HOL analysis is not approved for indexing. Read the methodology.
The current feed maps sgl-project/sglang (generic). Check affected ranges and fixed versions before updating.
| Package | Affected range | Fixed version |
|---|---|---|
| sgl-project/sglanggeneric | >=0.5.13 <=0.5.21 | Not reported |
Published upstream
Oct 16, 2026
Evidence: source:cvelist:source_dates:source-dates:recordSource modified
Oct 16, 2026
Evidence: source:cvelist:source_dates:source-dates:recordFirst seen by HOL
Oct 16, 2026
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.
Quoted source text, attributed separately from HOL analysis.