{
    "base_current_gpu_type": "NVIDIA A100-PCIE-40GB",
    "base_current_gpu_total_memory": 40339.3125,
    "base_token_generation_latency_sync": 55.742615127563475,
    "base_token_generation_latency_async": 56.15009795874357,
    "base_token_generation_throughput_sync": 0.01793959608302486,
    "base_token_generation_throughput_async": 0.01780940793255165,
    "base_token_generation_CO2_emissions": null,
    "base_token_generation_energy_consumption": null,
    "base_inference_latency_sync": 55.54227180480957,
    "base_inference_latency_async": 54.503512382507324,
    "base_inference_throughput_sync": 0.01800430496459108,
    "base_inference_throughput_async": 0.018347441408582427,
    "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": 174.30374908447266,
    "smashed_token_generation_latency_async": 175.11736061424017,
    "smashed_token_generation_throughput_sync": 0.005737111251206484,
    "smashed_token_generation_throughput_async": 0.005710456099226304,
    "smashed_token_generation_CO2_emissions": null,
    "smashed_token_generation_energy_consumption": null,
    "smashed_inference_latency_sync": 180.91202545166016,
    "smashed_inference_latency_async": 157.8697681427002,
    "smashed_inference_throughput_sync": 0.005527548528094949,
    "smashed_inference_throughput_async": 0.006334335014010341,
    "smashed_inference_CO2_emissions": null,
    "smashed_inference_energy_consumption": null
}