#!/bin/bash python -m llava.eval.model_vqa_loader \ --model-path liuhaotian/llava-v1.5-7b \ --question-file ./mme_pack/llava_mme_bench_replace.jsonl \ --image-folder ./mme_pack/MME_Benchmark_release_version \ --answers-file ./answers_hf_mme.jsonl \ --temperature 0 \ --conv-mode vicuna_v1
