{
    "base_current_gpu_type": "NVIDIA A100-PCIE-40GB",
    "base_current_gpu_total_memory": 40339.3125,
    "base_token_generation_latency_sync": 49.28324508666992,
    "base_token_generation_latency_async": 49.38601143658161,
    "base_token_generation_throughput_sync": 0.020290871638858027,
    "base_token_generation_throughput_async": 0.020248648775456117,
    "base_token_generation_CO2_emissions": null,
    "base_token_generation_energy_consumption": null,
    "base_inference_latency_sync": 48.311807250976564,
    "base_inference_latency_async": 46.43738269805908,
    "base_inference_throughput_sync": 0.020698873772307207,
    "base_inference_throughput_async": 0.02153437471922371,
    "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": 183.06483459472656,
    "smashed_token_generation_latency_async": 183.4463134407997,
    "smashed_token_generation_throughput_sync": 0.005462545563236242,
    "smashed_token_generation_throughput_async": 0.005451186133117424,
    "smashed_token_generation_CO2_emissions": null,
    "smashed_token_generation_energy_consumption": null,
    "smashed_inference_latency_sync": 190.26862182617188,
    "smashed_inference_latency_async": 149.299955368042,
    "smashed_inference_throughput_sync": 0.005255727352214666,
    "smashed_inference_throughput_async": 0.006697925645958045,
    "smashed_inference_CO2_emissions": null,
    "smashed_inference_energy_consumption": null
}