{
    "base_current_gpu_type": "NVIDIA A100-PCIE-40GB",
    "base_current_gpu_total_memory": 40339.3125,
    "base_token_generation_latency_sync": 74.98130340576172,
    "base_token_generation_latency_async": 74.85882211476564,
    "base_token_generation_throughput_sync": 0.013336658001108552,
    "base_token_generation_throughput_async": 0.013358478957455483,
    "base_token_generation_CO2_emissions": null,
    "base_token_generation_energy_consumption": null,
    "base_inference_latency_sync": 73.76199722290039,
    "base_inference_latency_async": 72.3219633102417,
    "base_inference_throughput_sync": 0.013557116640674919,
    "base_inference_throughput_async": 0.013827058257672983,
    "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": 196.6979736328125,
    "smashed_token_generation_latency_async": 192.5400961190462,
    "smashed_token_generation_throughput_sync": 0.005083936461220276,
    "smashed_token_generation_throughput_async": 0.005193723386227598,
    "smashed_token_generation_CO2_emissions": null,
    "smashed_token_generation_energy_consumption": null,
    "smashed_inference_latency_sync": 189.84580993652344,
    "smashed_inference_latency_async": 188.7110948562622,
    "smashed_inference_throughput_sync": 0.005267432556632977,
    "smashed_inference_throughput_async": 0.005299105496482237,
    "smashed_inference_CO2_emissions": null,
    "smashed_inference_energy_consumption": null
}