amd-strix-halo-vllm-toolboxes/benchmarks/benchmark_results/meta-llama_Meta-Llama-3.1-8B-Instruct_cluster_tp2_throughput.json

{
    "elapsed_time": 193.03236384499905,
    "num_requests": 100,
    "total_num_tokens": 74504,
    "requests_per_second": 0.5180478444552329,
    "tokens_per_second": 385.96636603292677
}