gshtras/vllm
A high-throughput and memory-efficient inference and serving engine for LLMs
Repositories
gshtras/scripts
gshtras/ci-infra
This repo hosts code for vLLM CI & Performance Benchmark infrastructure.
gshtras/pytorch
Tensors and Dynamic neural networks in Python with strong GPU acceleration