llama-3-1-8B-Instruct-GPTQ-Int4 Benchmark | LLM Performance Benchmarks