Tag: Paged Attention

• Supercharging Your Inference of Large Language Models with vLLM (part-2)

  As discussed in part 1 of this blog post vLLM is a high-throughput distributed system for serving large language models (LLMs) efficiently. It addresses the challenge of memory management in LLM serving systems by introducing PagedAttention, an innovative attention algorithm inspired by virtual memory techniques in operating systems. This approach allows for near-zero waste in…

  Alireza Darbehani

  August 10, 2024

  GenAI, Large Language Models, LLM Inference, MLOps

  Distributed Inference, GenAI, large-language-model, llm, llm-serving, Paged Attention

• Supercharging Your Inference of Large Language Models with vLLM (part-1)

  As the demand for large language models (LLMs) continues to rise, optimizing inference performance becomes crucial. vLLM is an innovative library designed to enhance the efficiency and speed of LLM inference and serving. This blog post explains a high level view of vLLM’s capabilities, its unique features, and how it compares to similar solutions in…

  Alireza Darbehani

  August 4, 2024

  GenAI, Large Language Models, LLM Inference

  Continuous Batching, GenAI, large-language-model, llm, LLM Inference, Paged Attention, vLLM