Cross-request KV cache exposure in the GLM-5.3 Workers AI model, and our remediation with SGLang

Details werden geladen...

https://blog.cloudflare.com/kv-cache-exposure-remediation/

Teilen bei Facebook
Teilen bei Twitter
Teilen bei Pinterest
Per Mail empfehlen

Cross-request KV cache exposure in the GLM-5.3 Workers AI model, and our remediation with SGLang | Cloudflare Blog

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.

Introducing Cloudflare Traces: follow requests through our entire platform | Cloudflare Blog

https://blog.cloudflare.com/cloudflare-tracing/

We tested our own WAF with frontier AI models. Here’s what we found | Cloudflare Blog

https://blog.cloudflare.com/adaptive-ai-waf-testing/

Registrieren

Cross-request KV cache exposure in the GLM-5.3 Workers AI model, and our remediation with SGLang | Cloudflare Blog

Ähnliche Seiten

Introducing Cloudflare Traces: follow requests through our entire platform | Cloudflare Blog

We tested our own WAF with frontier AI models. Here’s what we found | Cloudflare Blog