README.md

metadata

datasets:
  - Lin-Chen/ShareGPT4V
pipeline_tag: image-text-to-text
library_name: xtuner

Model

llava-llama-3-8b-v1_1 is a LLaVA model fine-tuned from meta-llama/Meta-Llama-3-8B-Instruct and CLIP-ViT-Large-patch14-336 with ShareGPT4V-PT and InternVL-SFT by XTuner.

Results

Model	MMBench Test (EN)	MMBench Test (CN)	CCBench Dev	MMMU Val	SEED-IMG	AI2D Test	ScienceQA Test	HallusionBench aAcc	POPE	GQA	TextVQA	MME	MMStar
LLaVA-v1.5-7B	66.5	59.0	27.5	35.3	60.5	54.8	70.4	44.9	85.9	62.0	58.2	1511/348	30.3
LLaVA-Llama-3-8B	68.9	61.6	30.4	36.8	69.8	60.9	73.3	47.3	87.2	63.5	58.0	1506/295	38.2
LLaVA-Llama-3-8B-v1.1	72.3	66.4	31.6	36.8	70.1	70.0	72.9	47.7	86.4	62.6	59.0	1469/349	45.1

Quickstart

Installation

pip install 'git+https://github.com/InternLM/xtuner.git#egg=xtuner[deepspeed]'

Chat

xtuner chat xtuner/llava-llama-3-8b-v1_1 \
  --visual-encoder openai/clip-vit-large-patch14-336 \
  --llava xtuner/llava-llama-3-8b-v1_1 \
  --prompt-template llama3_chat \
  --image $IMAGE_PATH

MMBench Evaluation

XTuner integrates the MMBench evaluation, and you can perform evaluations with the following command!

xtuner mmbench xtuner/llava-llama-3-8b-v1_1 \
  --visual-encoder openai/clip-vit-large-patch14-336 \
  --llava xtuner/llava-llama-3-8b-v1_1 \
  --prompt-template llama3_chat \
  --data-path $MMBENCH_DATA_PATH \
  --work-dir $RESULT_PATH

After the evaluation is completed, if it's a development set, it will directly print out the results; If it's a test set, you need to submit mmbench_result.xlsx to the official MMBench for final evaluation to obtain precision results!

Training

1. Pretrain (saved by default in ./work_dirs/llava_llama3_8b_instruct_clip_vit_large_p14_336_e1_gpu8_sharegpt4v_pretrain/)

NPROC_PER_NODE=8 xtuner train llava_llama3_8b_instruct_clip_vit_large_p14_336_e1_gpu8_sharegpt4v_pretrain --deepspeed deepspeed_zero2 --seed 1024

2. Fine-tune (saved by default in ./work_dirs/llava_llama3_8b_instruct_full_clip_vit_large_p14_336_lora_e1_gpu8_internvl_finetune/)

NPROC_PER_NODE=8 xtuner train llava_llama3_8b_instruct_full_clip_vit_large_p14_336_lora_e1_gpu8_internvl_finetune --deepspeed deepspeed_zero2 --seed 1024

Citation

@misc{2023xtuner,
    title={XTuner: A Toolkit for Efficiently Fine-tuning LLM},
    author={XTuner Contributors},
    howpublished = {\url{https://github.com/InternLM/xtuner}},
    year={2023}
}