{
    "base_current_gpu_type": "NVIDIA A100-PCIE-40GB",
    "base_current_gpu_total_memory": 40339.3125,
    "base_token_generation_latency_sync": 55.7013599395752,
    "base_token_generation_latency_async": 55.52884805947542,
    "base_token_generation_throughput_sync": 0.017952883037053304,
    "base_token_generation_throughput_async": 0.018008657390640042,
    "base_token_generation_CO2_emissions": null,
    "base_token_generation_energy_consumption": null,
    "base_inference_latency_sync": 55.23148803710937,
    "base_inference_latency_async": 54.08945083618164,
    "base_inference_throughput_sync": 0.01810561394485899,
    "base_inference_throughput_async": 0.018487893379222065,
    "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": 165.04326477050782,
    "smashed_token_generation_latency_async": 165.77455680817366,
    "smashed_token_generation_throughput_sync": 0.006059017321249051,
    "smashed_token_generation_throughput_async": 0.006032288785770375,
    "smashed_token_generation_CO2_emissions": null,
    "smashed_token_generation_energy_consumption": null,
    "smashed_inference_latency_sync": 171.79822235107423,
    "smashed_inference_latency_async": 129.57286834716797,
    "smashed_inference_throughput_sync": 0.005820781998293751,
    "smashed_inference_throughput_async": 0.007717665069516513,
    "smashed_inference_CO2_emissions": null,
    "smashed_inference_energy_consumption": null
}