{
    "base_current_gpu_type": "NVIDIA A100-PCIE-40GB",
    "base_current_gpu_total_memory": 40339.3125,
    "base_token_generation_latency_sync": 57.262949752807614,
    "base_token_generation_latency_async": 57.107177935540676,
    "base_token_generation_throughput_sync": 0.017463298770265843,
    "base_token_generation_throughput_async": 0.01751093358401886,
    "base_token_generation_CO2_emissions": null,
    "base_token_generation_energy_consumption": null,
    "base_inference_latency_sync": 56.82493438720703,
    "base_inference_latency_async": 55.75704574584961,
    "base_inference_throughput_sync": 0.017597908572774868,
    "base_inference_throughput_async": 0.017934953091994423,
    "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": 44.97442436218262,
    "smashed_token_generation_latency_async": 44.49597615748644,
    "smashed_token_generation_throughput_sync": 0.022234859349102958,
    "smashed_token_generation_throughput_async": 0.022473942283245094,
    "smashed_token_generation_CO2_emissions": null,
    "smashed_token_generation_energy_consumption": null,
    "smashed_inference_latency_sync": 51.584204864501956,
    "smashed_inference_latency_async": 42.89143085479736,
    "smashed_inference_throughput_sync": 0.019385779089291677,
    "smashed_inference_throughput_async": 0.02331468034688218,
    "smashed_inference_CO2_emissions": null,
    "smashed_inference_energy_consumption": null
}