amd-asalykov/vllm
A high-throughput and memory-efficient inference and serving engine for LLMs
Repositories
amd-asalykov/recipes
Common recipes to run vLLM
amd-asalykov/KernelBench
KernelBench: Can LLMs Write GPU Kernels? - Benchmark + Toolkit with Torch -> CUDA (+ more DSLs)