Collections:
  - Name: LLaVA
    Metadata:
      Architecture:
        - LLaMA
        - CLIP
    Paper:
      Title: Visual Instruction Tuning
      URL: https://arxiv.org/abs/2304.08485
    README: configs/llava/README.md

Models:
  - Name: llava-7b-v1_caption
    Metadata:
      FLOPs: null
      Parameters: 7045816320
    In Collection: LLaVA
    Results:
      - Task: Image Caption
        Dataset: COCO
        Metrics:
          BLEU-4: null
          CIDER: null
    Weights: https://download.openmmlab.com/mmclassification/v1/llava/llava-7b-v1_liuhaotian_20231025-c9e119b6.pth
    Config: configs/llava/llava-7b-v1_caption.py
  - Name: llava-7b-v1.5_caption
    Metadata:
      FLOPs: null
      Parameters: 7062900736
    In Collection: LLaVA
    Results:
      - Task: Image Caption
        Dataset: COCO
        Metrics:
          BLEU-4: null
          CIDER: null
    Weights: https://download.openmmlab.com/mmclassification/v1/llava/llava-7b-v1.5_liuhaotian_20231025-5828aa5a.pth
    Config: configs/llava/llava-7b-v1.5_caption.py
  - Name: llava-7b-v1.5_vqa
    Metadata:
      FLOPs: null
      Parameters: 7062900736
    In Collection: LLaVA
    Results:
      - Task: Visual Question Answering
        Dataset: COCO
        Metrics:
          BLEU-4: null
          CIDER: null
    Weights: https://download.openmmlab.com/mmclassification/v1/llava/llava-7b-v1.5_liuhaotian_20231025-5828aa5a.pth
    Config: configs/llava/llava-7b-v1.5_vqa.py