List your company

Efficiency and Optimization Evaluation on Multi-Objective Efficiency Evaluation

4.12LatencyGPTQ-INT4
3.85125.66567.489.2944Sep 18, 2025

Evaluation results

MethodLinks
GPTQ-INT4Model=OPT-6.7B2025.094.120.5970.488
FairGPTQ-INT4Model=OPT-6.7B2025.094.120.5890.481
GPTQ-INT4Model=Mistral-v0.3-7B2025.094.570.5840.478
FairGPTQ-INT4Model=Mistral-v0.3-7B2025.094.570.6070.497
GPTQ-INT4Model=Mistral-v0.3-7B-Instruct2025.094.570.5420.444
FairGPTQ-INT4Model=Mistral-v0.3-7B-Instruct2025.094.570.4030.331
GPTQ-INT4Model=Qwen-2.5-7B-Instruct2025.094.750.3670.304
FairGPTQ-INT4Model=Qwen-2.5-7B-Instruct2025.094.750.3660.304
GPTQ-INT4Model=LLaMA-3.1-8B-Instruct2025.095.020.6790.56
FairGPTQ-INT4Model=LLaMA-3.1-8B-Instruct2025.095.020.6010.497
GPTQ-INT4Model=Qwen-3-8B2025.095.340.3740.323
FairGPTQ-INT4Model=Qwen-3-8B2025.095.340.3620.313
Base-FP16Model=OPT-6.7B2025.099.750.580.677
Base-FP16Model=Qwen-2.5-7B-Instruct2025.099.840.3580.572
Base-FP16Model=Mistral-v0.3-7B2025.0910.250.5980.718
Base-FP16Model=Mistral-v0.3-7B-Instruct2025.0910.250.5260.68
Base-FP16Model=LLaMA-3.1-8B-Instruct2025.0910.730.5790.739
Base-FP16Model=Qwen-3-8B2025.0910.840.3840.657