{
    "base_current_gpu_type": "NVIDIA A100-PCIE-40GB",
    "base_current_gpu_total_memory": 40339.3125,
    "base_token_generation_latency_sync": 58.560171890258786,
    "base_token_generation_latency_async": 58.45238883048296,
    "base_token_generation_throughput_sync": 0.017076452607994227,
    "base_token_generation_throughput_async": 0.017107940667747342,
    "base_token_generation_CO2_emissions": null,
    "base_token_generation_energy_consumption": null,
    "base_inference_latency_sync": 55.90609893798828,
    "base_inference_latency_async": 55.28888702392578,
    "base_inference_throughput_sync": 0.017887136090629613,
    "base_inference_throughput_async": 0.01808681733034812,
    "base_inference_CO2_emissions": null,
    "base_inference_energy_consumption": null,
    "smashed_current_gpu_type": "NVIDIA A100-PCIE-40GB",
    "smashed_current_gpu_total_memory": 40339.3125,
    "smashed_token_generation_latency_sync": 175.9563995361328,
    "smashed_token_generation_latency_async": 174.2004169151187,
    "smashed_token_generation_throughput_sync": 0.005683226086895743,
    "smashed_token_generation_throughput_async": 0.005740514389740309,
    "smashed_token_generation_CO2_emissions": null,
    "smashed_token_generation_energy_consumption": null,
    "smashed_inference_latency_sync": 178.78282165527344,
    "smashed_inference_latency_async": 156.2723159790039,
    "smashed_inference_throughput_sync": 0.00559337855136992,
    "smashed_inference_throughput_async": 0.006399086068029834,
    "smashed_inference_CO2_emissions": null,
    "smashed_inference_energy_consumption": null
}