SGLang Developer Meeting: KV Cache Compression

​SGLang Dev Meeting: KV Cache Compression in SGLang

​For the latest session of the SGLang Dev Meeting, join us as Adrian Łańcucki (Senior Researcher @ NVIDIA) and Konrad Staniszewski (Senior Researcher @ NVIDIA) walk through their proposal for KV cache compression in SGLang.

​Liangsheng Yin (SGLang Core Dev) and Cheng Wan (SGLang Core Dev) join as panelists, followed by open discussion with everyone on the call.

​If you work on KV cache efficiency or long-context serving, or you just want a say in where SGLang goes next, come join. Bring your questions, ideas, and pushback.

​Join SGLang Slack 👉 https://slack.sglang.io/
Follow us on X 👉 /sgl_project

​If SGLang helps you, consider giving us a star on GitHub. It really motivates the team. ⭐ /sgl-project/sglang