pip install -r requirements.txt# eval multiple models on multiple benchmarks
bash submit_eval.sh "OCRBench MMStar MMMU_DEV_VAL ChartQA_TEST DocVQA_VAL HallusionBench ScienceQA_TEST MathVista_MINI TextVQA_VAL AI2D_TEST InfoVQA_VAL RealWorldQA MMVet MME" "Eagle2-1B Eagle2-2B" work_dirs/eagel2-1b test 8
bash submit_eval.sh "OCRBench" "Eagle2-1B Eagle2-2B" work_dirs/eagel2-1b test 8