{
    "base_current_gpu_type": "NVIDIA A100-PCIE-40GB",
    "base_current_gpu_total_memory": 40339.3125,
    "base_token_generation_latency_sync": 55.14786720275879,
    "base_token_generation_latency_async": 55.07155656814575,
    "base_token_generation_throughput_sync": 0.018133067527767867,
    "base_token_generation_throughput_async": 0.01815819385389255,
    "base_token_generation_CO2_emissions": null,
    "base_token_generation_energy_consumption": null,
    "base_inference_latency_sync": 55.21827850341797,
    "base_inference_latency_async": 54.163432121276855,
    "base_inference_throughput_sync": 0.018109945241014728,
    "base_inference_throughput_async": 0.01846264095231094,
    "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": 173.5607147216797,
    "smashed_token_generation_latency_async": 174.04278870671988,
    "smashed_token_generation_throughput_sync": 0.005761672516753521,
    "smashed_token_generation_throughput_async": 0.00574571349626616,
    "smashed_token_generation_CO2_emissions": null,
    "smashed_token_generation_energy_consumption": null,
    "smashed_inference_latency_sync": 179.38524017333984,
    "smashed_inference_latency_async": 156.71169757843018,
    "smashed_inference_throughput_sync": 0.00557459464911216,
    "smashed_inference_throughput_async": 0.00638114458239166,
    "smashed_inference_CO2_emissions": null,
    "smashed_inference_energy_consumption": null
}