amd-strix-halo-vllm-toolboxes/benchmarks/benchmark_results/meta-llama_Meta-Llama-3.1-8B-Instruct_tp1_throughput.json

{
    "elapsed_time": 311.826995067001,
    "num_requests": 100,
    "total_num_tokens": 74504,
    "requests_per_second": 0.3206906444341466,
    "tokens_per_second": 238.92735772921657
}