diff --git a/scripts/multimodal/run_finetune_multi_modal_stage2_zero3.sh b/scripts/multimodal/run_finetune_multi_modal_stage2_zero3.sh
deleted file mode 100755
index 4cce676d7..000000000
--- a/scripts/multimodal/run_finetune_multi_modal_stage2_zero3.sh
+++ /dev/null
@@ -1,91 +0,0 @@
-#!/bin/bash
-# Please run this script under ${project_id} in project directory of
-#   https://github.com/shizhediao/llm-ft
-#     COMMIT: d5fecf30ba8011067b10cf51fede53a5ab6574e4
-
-# Parses argumen
-# FIXME upload the projector into lmflow.
-model_name_or_path=Salesforce/blip2-flan-t5-xxl
-# Please download the coco dataset and the conversation file from
-# https://huggingface.co/datasets/liuhaotian/LLaVA-Instruct-150K/blob/main/llava_instruct_80k.json
-dataset_path=./data/llava_instruct_80k_truncated.json
-image_folder=./data/coco2017/train2017
-output_dir=output_models/finetune_llava-336px-vicuna-7b-v1.3
-deepspeed_args="--master_port=12300 --include localhost:7"
-
-while [[ $# -ge 1 ]]; do
-  key="$1"
-  case ${key} in
-    -m|--model_name_or_path)
-      model_name_or_path="$2"
-      shift
-      ;;
-    -d|--dataset_path)
-      dataset_path="$2"
-      shift
-      ;;
-    -o|--output_model_path)
-      output_dir="$2"
-      shift
-      ;;
-    --deepspeed_args)
-      deepspeed_args="$2"
-      shift
-      ;;
-    *)
-      echo "error: unknown option \"${key}\"" 1>&2
-      exit 1
-  esac
-  shift
-done
-
-if [ ! -f output_models/llava-336px-pretrain-vicuna-7b-v1.3_language_projection.pth ]; then
-  cd output_models && ./download.sh llava_vicuna7b_language_projection && cd -
-fi
-
-if [ ! -d data/coco2017 ]; then
-  cd data && ./download.sh coco2017 && cd -
-fi
-
-if [ ! -f data/llava_instruct_80k.json ]; then
-  cd data && ./download.sh llava_instruction_finetune_80k && cd -
-fi
-# Finetune
-exp_id=finetune
-project_dir=$(cd "$(dirname $0)"/..; pwd)
-log_dir=${project_dir}/log/${exp_id}
-mkdir -p ${output_dir} ${log_dir}
-# train batch size is set as 4
-# default in llava is 16
-deepspeed ${deepspeed_args} \
-  examples/finetune_multi_modal.py \
-    --deepspeed configs/ds_config_zero3.json \
-    --arch_type vision_encoder_decoder \
-    --llava_loading True \
-    --model_name_or_path ${model_name_or_path} \
-    --image_encoder_name_or_path openai/clip-vit-large-patch14-336 \
-    --pretrained_language_projection_path output_models/llava-336px-pretrain-vicuna-7b-v1.3_language_projection.pth \
-    --dataset_path ${dataset_path} \
-    --output_dir ${output_dir} --overwrite_output_dir \
-    --image_folder ${image_folder} \
-    --custom_vision_model True \
-    --llm_model_name_or_path lmsys/vicuna-7b-v1.3 \
-    --image_aspect_ratio None \
-    --learning_rate 2e-5 \
-    --gradient_accumulation_steps 8 \
-    --per_device_train_batch_size 4 \
-    --per_device_eval_batch_size 4 \
-    --weight_decay 0. \
-    --warmup_ratio 0.03 \
-    --lr_scheduler_type "cosine" \
-    --run_name finetune \
-    --validation_split_percentage 0 \
-    --logging_steps 20 \
-    --do_train \
-    --ddp_timeout 72000 \
-    --save_steps 5000 \
-    --dataloader_num_workers 1 \
-    --sep_style "v1" \
-    --num_train_epochs 1 \
-    | tee ${log_dir}/train.log \
-    2> ${log_dir}/train.err