{
    "base_current_gpu_type": "NVIDIA A100-PCIE-40GB",
    "base_current_gpu_total_memory": 40339.3125,
    "base_token_generation_latency_sync": 48.19650077819824,
    "base_token_generation_latency_async": 47.3951056599617,
    "base_token_generation_throughput_sync": 0.02074839425795724,
    "base_token_generation_throughput_async": 0.021099225037592376,
    "base_token_generation_CO2_emissions": null,
    "base_token_generation_energy_consumption": null,
    "base_inference_latency_sync": 46.33999366760254,
    "base_inference_latency_async": 44.7718620300293,
    "base_inference_throughput_sync": 0.021579631779257778,
    "base_inference_throughput_async": 0.022335457018278174,
    "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": 184.61887969970704,
    "smashed_token_generation_latency_async": 184.8322255536914,
    "smashed_token_generation_throughput_sync": 0.005416564121863138,
    "smashed_token_generation_throughput_async": 0.005410311957259384,
    "smashed_token_generation_CO2_emissions": null,
    "smashed_token_generation_energy_consumption": null,
    "smashed_inference_latency_sync": 190.98859558105468,
    "smashed_inference_latency_async": 150.1239776611328,
    "smashed_inference_throughput_sync": 0.005235914725471681,
    "smashed_inference_throughput_async": 0.006661161098843577,
    "smashed_inference_CO2_emissions": null,
    "smashed_inference_energy_consumption": null
}