Skip to content

Datasets

Hugging Face datasets with plain-language descriptions: what's inside and what models to train on them. 4,788 in the catalog so far, with new ones added every hour.

Data type

For which models

Find a dataset for your goal

Answered by Gemini 4 · the chat is kept only in this tab

DatasetZenoNing/GarmentCodeVTONDataset

Synthetic Virtual Try-On Renders

Generated images of people in various outfits for virtual try-on tasks.

  • Image → image
50.9K 310K–100K rows
DatasetYuan-Shenghai/MMAUDTop pick

Multimodal Drone Detection Dataset

Camera, lidar, radar, and microphone recordings for detecting and tracking miniature UAVs.

  • English
50.6K 310K–100K rows
Datasetmhjiang0408/MAC_Bench

Scientific cover benchmark for multimodal LLMs

A dynamic English-language benchmark for assessing how well multimodal LLMs understand scientific journals and illustrations.

  • Image text → text
  • English
50.5K 11K–10K rows
Datasetphysical-intelligence/libero

LIBERO Robots in LeRobot Format

Four LIBERO datasets with Panda robot episodes, merged and converted to the LeRobot format.

  • Robotics
50.4K 98100K–1M rows
DatasetVoxel51/GQA-Scene-Graph

GQA sample of 35k images with scene graphs

35,000 images from the GQA dataset in FiftyOne format with object and attribute scene graph annotations, but without questions.

  • Object detection
  • English
49.7K 410K–100K rows
DatasetGreenNode/GreenOCR-Dataset

Synthetic document pages for OCR

Generated pages in Vietnamese and English with block annotations for training document parsing.

  • Image → text
  • Image text → text
  • Vietnamese
  • English
49.4K 11K–10K rows🔒 on request
Datasethaifan-gong/TN3K

Annotated thyroid ultrasound nodule images (TN3K)

3,493 medical ultrasound images of the thyroid with high-quality annotations for segmentation and classification tasks.

  • Image segmentation
  • Image classification
49.3K 11K–10K rows
Datasettiagozip/vkTop pick

Multi-Billion VK Public Data Scrap

Regularly updated corpus of posts, comments, photos, videos, and profiles from VK in Russian and Ukrainian.

  • Text classification
  • Russian
  • Ukrainian
49.1K 01B–10B rows🔒 on request
Datasetlmms-lab-encoder/textvqa

TextVQA Image Text Reading Benchmark

Formatted version of the TextVQA benchmark for evaluating multimodal models in lmms-eval.

    48.8K 3010K–100K rows
    Datasetmixed-modality-search/MixBench25

    Multimodal Text and Image Search Benchmark

    Set of queries and mixed corpora based on MSCOCO, Google_WIT, VisualNews, and OVEN for cross-modal search evaluation.

    • Text ranking
    • English
    48.7K 0
    Datasetllamaindex/ParseBench

    ParseBench benchmark for document parsing

    A benchmark for evaluating real enterprise document parsing systems across five dimensions.

    • English
    48.4K 173100K–1M rows
    DatasetFINEBENCH/FineBench

    FineBench for video VQA

    A large-scale multimodal benchmark for evaluating action understanding in video.

    • Visual question answering
    • Video text → text
    • English
    48.3K 1100K–1M rows
    DatasetAmazonScience/document-haystack

    Long Document Analysis Benchmark for Visual LLMs

    Multimodal dataset from Amazon Science to test VLM capability to find answers in bulky documents and images.

    • Question answering
    • Visual question answering
    • Document question answering
    • English
    47.6K 26
    Datasetbiraripc/curated_mnist_export

    Curated MNIST dataset

    A processed and combined set of MNIST images.

      47.4K 010K–100K rows
      Datasetopendatalab/OmniDocBenchTop pick

      OmniDocBench: Document parsing benchmark

      A benchmark of 1,651 PDF pages for evaluating document parsing across various types and languages.

        46.4K 1241K–10K rows
        DatasetAI4Math/MathVista

        MathVista: Visual-mathematical benchmark

        Mathematical reasoning benchmark over images: IQTest, FunctionQA, and PaperQA; languages en, zh, fa.

        • Multiple choice
        • Question answering
        • Visual question answering
        • English
        • Chinese
        • Persian
        45.9K 2371K–10K rows
        DatasetGoku-OpenLab/gpt-image-2-prompts-datasets

        Prompts and generations for GPT Image 2

        Over 15 thousand English and Chinese prompts with metadata and generated images from GPT Image 2.

        • Text → image
        • English
        • Chinese
        45.8K 2210K–100K rows