vLLM: High-performance serving of LLMs using open-source technology

vLLM: High-performance serving of LLMs using open-source technology
Share:

Download MP3


Similar Tracks

Learning from Meta Llama2 & Llama3 Training Environments: Best Practices in AI Storage Architecture AI Infra Forum
How can GPUaaS providers improve their ROI by offering on-demand instances and PaaS functionality AI Infra Forum
Measuring AI: Why benchmarks matter, and how to build the right ones. AI Infra Forum
Accelerating LLM Inference with vLLM Databricks
Hive Thrift to Apache Arrow 10x BI connectivity IOMETE
Transformers (how LLMs work) explained visually | DL5 3Blue1Brown
Networking in the Age of AI AI Infra Forum
UnifabriX Memory over Fabrics: An Open Journey from CXL to UALink in AI Infrastructures AI Infra Forum
The Memory Wall in AI - A Crisis we must solve AI Infra Forum
Enlightening the path forward for massive-scale! Laser-Focused! Through Memory Lanes! AI Infra Forum
Ultra Accelerator Link (UAL): Accelerator Scale-up network AI Infra Forum
GenAI meets Storage: RAG is the New Killer App AI Infra Forum
Next Gen Inference for Fine-tuned LLMs - Blazing Fast & Cost-Effective Predibase
CXL Market Perspective: Get Ready for Adoption and Growth in 2025 AI Infra Forum
Delivering your own state-of-the-art enterprise LLMs AI Infra Forum
vLLM Office Hours - Distributed Inference with vLLM - January 23, 2025 Neural Magic
RAG vs. Fine Tuning IBM Technology
Contributing To Open Source – Beginner's Guide freeCodeCamp.org
Revolutionizing GPU-as-a-Service for Maximum Efficiency AI Infra Forum
vLLM: Easy, Fast, and Cheap LLM Serving for Everyone - Woosuk Kwon & Xiaoxuan Liu, UC Berkeley PyTorch