amd-strix-halo-vllm-toolboxes/benchmarks/benchmark_results/btbtyler09_Qwen3-Coder-30B-A3B-Instruct-gptq-4bit_cluster_tp2_throughput.json

{
    "elapsed_time": 442.1101265470061,
    "num_requests": 100,
    "total_num_tokens": 75285,
    "requests_per_second": 0.2261879880043141,
    "tokens_per_second": 170.28562676904787
}