Llama3-8b-instruct-fp16.gguf Benchmark | LLM Performance Benchmarks