Files
amd-strix-halo-vllm-toolboxes/benchmarks/benchmark_results_rocm/btbtyler09_Qwen3-Coder-30B-A3B-Instruct-gptq-8bit_tp1_throughput.json
T

7 lines
189 B
JSON

{
"elapsed_time": 838.6815635629973,
"num_requests": 100,
"total_num_tokens": 75285,
"requests_per_second": 0.11923476602390883,
"tokens_per_second": 89.76589360109976
}