{
    "base_current_gpu_type": "NVIDIA A100-PCIE-40GB",
    "base_current_gpu_total_memory": 40339.3125,
    "base_token_generation_latency_sync": 58.130772018432616,
    "base_token_generation_latency_async": 57.485764659941196,
    "base_token_generation_throughput_sync": 0.017202592796856563,
    "base_token_generation_throughput_async": 0.017395610998923484,
    "base_token_generation_CO2_emissions": null,
    "base_token_generation_energy_consumption": null,
    "base_inference_latency_sync": 55.49096908569336,
    "base_inference_latency_async": 55.28078079223633,
    "base_inference_throughput_sync": 0.01802095037222587,
    "base_inference_throughput_async": 0.01808946953477981,
    "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": 163.0220962524414,
    "smashed_token_generation_latency_async": 163.83853014558554,
    "smashed_token_generation_throughput_sync": 0.006134137782472688,
    "smashed_token_generation_throughput_async": 0.0061035703818351425,
    "smashed_token_generation_CO2_emissions": null,
    "smashed_token_generation_energy_consumption": null,
    "smashed_inference_latency_sync": 170.61754913330077,
    "smashed_inference_latency_async": 128.1928300857544,
    "smashed_inference_throughput_sync": 0.005861061802140388,
    "smashed_inference_throughput_async": 0.007800748289362607,
    "smashed_inference_CO2_emissions": null,
    "smashed_inference_energy_consumption": null
}