MQT-LLaVA-7b / README.md
gordonhubackup's picture
init
089bfb3
metadata
inference: false
pipeline_tag: image-text-to-text


MQT-LLaVA Model Card

Model details

Model type: MQT-LLaVA is an open-source chatbot trained by fine-tuning LLaMA/Vicuna on GPT-generated multimodal instruction-following data. It is an auto-regressive language model, based on the transformer architecture.

Model date: MQT-LLaVA-7B was trained in May 2024. Paper

License

Llama 2 is licensed under the LLAMA 2 Community License, Copyright (c) Meta Platforms, Inc. All Rights Reserved.

Where to send questions or comments about the model: https://github.com/gordonhu608/MQT-LLaVA/issues

Intended use

Primary intended uses: The primary use of MQT-LLaVA is research on large multimodal models and chatbots.

Primary intended users: The primary intended users of the model are researchers and hobbyists in computer vision, natural language processing, machine learning, and artificial intelligence.

Training dataset

  • 558K filtered image-text pairs from LAION/CC/SBU, captioned by BLIP.
  • 158K GPT-generated multimodal instruction-following data.
  • 450K academic-task-oriented VQA data mixture.
  • 40K ShareGPT data.

Evaluation dataset

A collection of 11 benchmarks, including 4 academic VQA benchmarks and 7 recent benchmarks specifically proposed for instruction-following LMMs.