vLLM: Easy, Fast, and Cheap LLM Serving for Everyone

​For developers, builders, AI enthusiasts, and anyone looking to optimize LLM serving and opportunity to contribute to open source project.

​
📅 When: July 18th, 2024

​⏰ Time: 10:30 AM PST

​📍 Where: Zoom Webinar

​
📝 Agenda:

​This presentation introduces vLLM, a high-performance open-source LLM inference engine. We'll cover:

• ​Overview of vLLM and its key benefits

• ​Deep dive into features like pipeline parallelism and speculative decoding

• ​Live demonstration of setting up and optimizing LLM inference

• ​Contribution opportunities and current development priorities

​
🎤 Speakers:

• ​Woosuk Kwon - Ph.D. student at UC Berkeley, creator of vLLM

• ​Kaichao You - Ph.D. student at Tsinghua University, vLLM contributor
  

​💫 What's the Deal?

​We promise 1 hour of cutting-edge AI technology, insightful demonstrations, and the chance to connect with the creators of a amazing open-source project. Learn how to significantly improve your AI infrastructure's performance and cost-efficiency!

​🔗 Resources:

• ​vLLM GitHub: https://github.com/vllm-project/vllm