mindef-overdracht/llm-throughput-tests-mindef-metadateren/results/results_openai-gpt-oss-120b-max-16/benchmark_results.json
2026-06-02 11:46:24 +02:00

58 lines
1.4 KiB
JSON

{
"timestamp": "2026-03-17T09:42:27.751665",
"model_name": "openai-gpt-oss-120b-max-16",
"results": [
{
"config": {
"input_tokens": 50000,
"output_tokens": 1024,
"batch_size": 16,
"num_batches": 1,
"total_requests": 16,
"actual_input_tokens": 40691
},
"success_metrics": {
"success_rate": 100.0,
"successful_requests": 16,
"failed_requests": 0
},
"latency": {
"mean": 105.895,
"std": 0.51,
"min": 104.893,
"max": 106.744,
"p50": 105.983,
"p95": 106.564,
"p99": 106.708,
"ci_95_lower": 105.645,
"ci_95_upper": 106.144
},
"ttft": {
"mean": 73.976,
"std": 2.6,
"p50": 73.007,
"p90": 76.235
},
"tokens": {
"total_generated": 16384,
"content_tokens": 14467,
"reasoning_tokens": 1917,
"avg_per_request": 1024.0
},
"throughput": {
"concurrent_total_tps": 153.35,
"concurrent_content_tps": 135.41,
"requests_per_second": 0.15,
"actual_wall_time": 106.84,
"efficiency_percent": 99.11
},
"batch_metrics": {
"num_batches": 1,
"avg_batch_size": 16.0,
"avg_batch_throughput": 153.35,
"min_batch_throughput": 153.35,
"max_batch_throughput": 153.35
}
}
]
}