Story
cloudflare_blog · Oct 26, 2026 · news
Source brief
Cross-request KV cache exposure in the GLM-5.3 Workers AI model, and our remediation with SGLang
blog.cloudflare.comOct 26, 2026
original source linked
In brief
A configuration mismatch in GLM 5.3 on Workers AI triggered an SGLang bug that exposed KV cache pages across requests. Here is how we addressed the issue.
