Add vLLM to local-apps

[mgoin]

vLLM is a high-throughput and memory-efficient open-source serving engine for LLMs.

vLLM is fast with:

• State-of-the-art serving throughput
• Efficient management of attention key and value memory with PagedAttention
• Continuous batching of incoming requests
• Fast model execution with CUDA/HIP graph
• Quantization: FP8, GPTQ, AWQ, SqueezeLLM, FP8 KV Cache
• Optimized CUDA kernels

Transparent Logo: