amd-strix-halo-vllm-toolboxes/benchmarks/benchmark_results/openai_gpt-oss-20b_cluster_tp2_throughput.json

{
    "elapsed_time": 271.0714236530039,
    "num_requests": 100,
    "total_num_tokens": 74946,
    "requests_per_second": 0.36890646255655896,
    "tokens_per_second": 276.48063742763867
}