{
    "base_current_gpu_type": "NVIDIA A100-PCIE-40GB",
    "base_current_gpu_total_memory": 40339.3125,
    "base_token_generation_latency_sync": 68.76320419311523,
    "base_token_generation_latency_async": 68.87206453830004,
    "base_token_generation_throughput_sync": 0.01454266146748471,
    "base_token_generation_throughput_async": 0.01451967509183489,
    "base_token_generation_CO2_emissions": null,
    "base_token_generation_energy_consumption": null,
    "base_inference_latency_sync": 64.66969566345215,
    "base_inference_latency_async": 63.82594108581543,
    "base_inference_throughput_sync": 0.015463193227382798,
    "base_inference_throughput_async": 0.015667610739267868,
    "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": 51.486428833007814,
    "smashed_token_generation_latency_async": 52.597666531801224,
    "smashed_token_generation_throughput_sync": 0.0194225939274876,
    "smashed_token_generation_throughput_async": 0.019012250275311875,
    "smashed_token_generation_CO2_emissions": null,
    "smashed_token_generation_energy_consumption": null,
    "smashed_inference_latency_sync": 54.388633728027344,
    "smashed_inference_latency_async": 55.66208362579346,
    "smashed_inference_throughput_sync": 0.018386194531021725,
    "smashed_inference_throughput_async": 0.01796555096145568,
    "smashed_inference_CO2_emissions": null,
    "smashed_inference_energy_consumption": null
}