{
    "base_current_gpu_type": "NVIDIA A100-PCIE-40GB",
    "base_current_gpu_total_memory": 40339.3125,
    "base_token_generation_latency_sync": 74.54675521850587,
    "base_token_generation_latency_async": 74.43354055285454,
    "base_token_generation_throughput_sync": 0.013414400091176,
    "base_token_generation_throughput_async": 0.013434803619074249,
    "base_token_generation_CO2_emissions": null,
    "base_token_generation_energy_consumption": null,
    "base_inference_latency_sync": 73.78821182250977,
    "base_inference_latency_async": 72.89724349975586,
    "base_inference_throughput_sync": 0.013552300229275118,
    "base_inference_throughput_async": 0.01371793982859378,
    "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": 188.3400436401367,
    "smashed_token_generation_latency_async": 188.5381616652012,
    "smashed_token_generation_throughput_sync": 0.0053095453344521384,
    "smashed_token_generation_throughput_async": 0.005303966004377201,
    "smashed_token_generation_CO2_emissions": null,
    "smashed_token_generation_energy_consumption": null,
    "smashed_inference_latency_sync": 186.75937194824218,
    "smashed_inference_latency_async": 183.61716270446777,
    "smashed_inference_throughput_sync": 0.005354483630824891,
    "smashed_inference_throughput_async": 0.005446113997576045,
    "smashed_inference_CO2_emissions": null,
    "smashed_inference_energy_consumption": null
}