{
    "base_current_gpu_type": "NVIDIA A100-PCIE-40GB",
    "base_current_gpu_total_memory": 40339.3125,
    "base_token_generation_latency_sync": 66.09409217834472,
    "base_token_generation_latency_async": 65.97807370126247,
    "base_token_generation_throughput_sync": 0.015129945310416762,
    "base_token_generation_throughput_async": 0.015156550409880567,
    "base_token_generation_CO2_emissions": null,
    "base_token_generation_energy_consumption": null,
    "base_inference_latency_sync": 65.53385009765626,
    "base_inference_latency_async": 63.47980499267578,
    "base_inference_throughput_sync": 0.015259289642067955,
    "base_inference_throughput_async": 0.015753041461223437,
    "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": 43.21300010681152,
    "smashed_token_generation_latency_async": 43.23582127690315,
    "smashed_token_generation_throughput_sync": 0.02314118430861673,
    "smashed_token_generation_throughput_async": 0.02312896969379893,
    "smashed_token_generation_CO2_emissions": null,
    "smashed_token_generation_energy_consumption": null,
    "smashed_inference_latency_sync": 60.904960250854494,
    "smashed_inference_latency_async": 42.44384765625,
    "smashed_inference_throughput_sync": 0.01641902393304608,
    "smashed_inference_throughput_async": 0.023560540696002302,
    "smashed_inference_CO2_emissions": null,
    "smashed_inference_energy_consumption": null
}