THREAT OPS › Threat News › Cross-request KV cache exposure in the GLM-5.3 Workers AI model, and our remediation with SGLang
Cross-request KV cache exposure in the GLM-5.3 Workers AI model, and our remediation with SGLang
<p>On September 8, 2026, we were alerted by a security researcher, Gadi Evron from Knostic, to a tenant isolation issue affecting the GLM-5.3 Workers AI model (<code>@cf/zai-org/glm-5.3</code>) starting on August 27, 2026. We have remediated the issue, and we restored the model on September 9. We have confirmed that no other model hosted on Workers AI was affected.</p><p>Between August 27 and Sept
Indicators of compromise
- http://z.aiurl
- https://www.cloudflarestatus.com/incidents/jsvvrtk5hvymurl
- https://kvcache-ai.github.io/Mooncake/url
- https://www.lmsys.org/blog/2026-10-26-cf-securityurl
- https://hackerone.com/cloudflareurl
Original source: https://blog.cloudflare.com/kv-cache-exposure-remediation/