amd-strix-halo-vllm-toolboxes/benchmarks/benchmark_results/btbtyler09_Qwen3-Coder-30B-A3B-Instruct-gptq-8bit_cluster_tp2_throughput.json

{
    "elapsed_time": 577.3050836349939,
    "num_requests": 100,
    "total_num_tokens": 75285,
    "requests_per_second": 0.1732186374842766,
    "tokens_per_second": 130.40765123003763
}