{
    "base_current_gpu_type": "NVIDIA A100-PCIE-40GB",
    "base_current_gpu_total_memory": 40339.3125,
    "base_token_generation_latency_sync": 75.25102386474609,
    "base_token_generation_latency_async": 75.25082267820835,
    "base_token_generation_throughput_sync": 0.013288855734340169,
    "base_token_generation_throughput_async": 0.013288891262707575,
    "base_token_generation_CO2_emissions": null,
    "base_token_generation_energy_consumption": null,
    "base_inference_latency_sync": 72.9151481628418,
    "base_inference_latency_async": 72.50525951385498,
    "base_inference_throughput_sync": 0.013714571322912142,
    "base_inference_throughput_async": 0.013792102899913223,
    "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": 163.4791748046875,
    "smashed_token_generation_latency_async": 164.00436963886023,
    "smashed_token_generation_throughput_sync": 0.00611698707920886,
    "smashed_token_generation_throughput_async": 0.006097398515673778,
    "smashed_token_generation_CO2_emissions": null,
    "smashed_token_generation_energy_consumption": null,
    "smashed_inference_latency_sync": 170.5095184326172,
    "smashed_inference_latency_async": 131.371808052063,
    "smashed_inference_throughput_sync": 0.005864775228927675,
    "smashed_inference_throughput_async": 0.007611983231620725,
    "smashed_inference_CO2_emissions": null,
    "smashed_inference_energy_consumption": null
}