{
    "base_current_gpu_type": "NVIDIA A100-PCIE-40GB",
    "base_current_gpu_total_memory": 40339.3125,
    "base_token_generation_latency_sync": 48.8967472076416,
    "base_token_generation_latency_async": 48.937406204640865,
    "base_token_generation_throughput_sync": 0.020451258153297354,
    "base_token_generation_throughput_async": 0.02043426649582354,
    "base_token_generation_CO2_emissions": null,
    "base_token_generation_energy_consumption": null,
    "base_inference_latency_sync": 48.2050048828125,
    "base_inference_latency_async": 46.3212251663208,
    "base_inference_throughput_sync": 0.0207447339219449,
    "base_inference_throughput_async": 0.021588375445800584,
    "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": 186.9020736694336,
    "smashed_token_generation_latency_async": 187.87380326539278,
    "smashed_token_generation_throughput_sync": 0.005350395425620911,
    "smashed_token_generation_throughput_async": 0.00532272186233111,
    "smashed_token_generation_CO2_emissions": null,
    "smashed_token_generation_energy_consumption": null,
    "smashed_inference_latency_sync": 193.66358947753906,
    "smashed_inference_latency_async": 165.66784381866455,
    "smashed_inference_throughput_sync": 0.005163593232459317,
    "smashed_inference_throughput_async": 0.006036174413512452,
    "smashed_inference_CO2_emissions": null,
    "smashed_inference_energy_consumption": null
}