|
{ |
|
"base_current_gpu_type": "NVIDIA A100-PCIE-40GB", |
|
"base_current_gpu_total_memory": 40339.3125, |
|
"base_token_generation_latency_sync": 57.080553436279295, |
|
"base_token_generation_latency_async": 56.89493417739868, |
|
"base_token_generation_throughput_sync": 0.017519101336610717, |
|
"base_token_generation_throughput_async": 0.01757625726188548, |
|
"base_token_generation_CO2_emissions": null, |
|
"base_token_generation_energy_consumption": null, |
|
"base_inference_latency_sync": 55.81363220214844, |
|
"base_inference_latency_async": 54.92098331451416, |
|
"base_inference_throughput_sync": 0.017916769802369302, |
|
"base_inference_throughput_async": 0.01820797698164531, |
|
"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": 167.02207641601564, |
|
"smashed_token_generation_latency_async": 168.5217283666134, |
|
"smashed_token_generation_throughput_sync": 0.0059872324752400854, |
|
"smashed_token_generation_throughput_async": 0.005933952907393244, |
|
"smashed_token_generation_CO2_emissions": null, |
|
"smashed_token_generation_energy_consumption": null, |
|
"smashed_inference_latency_sync": 174.03873138427736, |
|
"smashed_inference_latency_async": 141.34645462036133, |
|
"smashed_inference_throughput_sync": 0.00574584744468173, |
|
"smashed_inference_throughput_async": 0.007074814877287678, |
|
"smashed_inference_CO2_emissions": null, |
|
"smashed_inference_energy_consumption": null |
|
} |