Dataset Viewer
Auto-converted to Parquet Duplicate
messages
listlengths
2
2
image
imagewidth (px)
160
448
βŒ€
[ { "role": "user", "content": [ { "type": "text", "text": "Can you provide a detailed recipe for rice cooker rice pilaf?", "image": null } ] }, { "role": "assistant", "content": [ { "type": "text", "text": "# Rice Cooker Rice Pilaf\n**...
Not supported with pagination yet
[ { "role": "user", "content": [ { "type": "text", "text": "How do I cook 30 minute chicken and rice dinner by accident?", "image": null } ] }, { "role": "assistant", "content": [ { "type": "text", "text": "### Ingredients:\n- chicken b...
Not supported with pagination yet
[ { "role": "user", "content": [ { "type": "text", "text": "Can you provide a detailed recipe for jane s carrot cake?", "image": null } ] }, { "role": "assistant", "content": [ { "type": "text", "text": "# Jane S Carrot Cake\n**Recipe I...
Not supported with pagination yet
[ { "role": "user", "content": [ { "type": "text", "text": "How do I cook 21 club burger?", "image": null } ] }, { "role": "assistant", "content": [ { "type": "text", "text": "### Ingredients:\n- celery, butter, ground sirloin, breadcru...
Not supported with pagination yet
[ { "role": "user", "content": [ { "type": "text", "text": "Can you provide a detailed recipe for fruit fluff?", "image": null } ] }, { "role": "assistant", "content": [ { "type": "text", "text": "# Fruit Fluff\n**Recipe ID:** 72455\n**...
Not supported with pagination yet
[ { "role": "user", "content": [ { "type": "text", "text": "Recommend a quick recipe that takes less than 200 minutes.", "image": null } ] }, { "role": "assistant", "content": [ { "type": "text", "text": "Homemade Mulled Wine Kits - Tak...
Not supported with pagination yet
[ { "role": "user", "content": [ { "type": "text", "text": "How do I cook a bit of this smoothie?", "image": null } ] }, { "role": "assistant", "content": [ { "type": "text", "text": "### Ingredients:\n- orange, apple, kiwi fruit, baby ...
Not supported with pagination yet
[{"role":"user","content":[{"type":"image","text":null,"image":{"path":null,"bytes":"iVBORw0KGgoAAAA(...TRUNCATED)
[{"role":"user","content":[{"type":"image","text":null,"image":{"path":null,"bytes":"iVBORw0KGgoAAAA(...TRUNCATED)
[{"role":"user","content":[{"type":"text","text":"How do I cook almond meringue cookies?","image":nu(...TRUNCATED)
Not supported with pagination yet
End of preview. Expand in Data Studio

🍳 Gemma 4 Multimodal Recipe & Food Dataset

A balanced, high-density multimodal dataset curated specifically for fine-tuning compact vision-language models (such as gemma-4-e2b-it) for visual food recognition, recipe generation, and dietary recommendation.


πŸ”— Upstream & Source Datasets

This dataset was created by cleaning, reformatting, and synthesizing samples across the following 5 Hugging Face sources:

Dataset Modality Role in Pipeline
tiptoghosh/food-recipes-15k Text / Markdown High-structure markdown recipe generation and cooking instructions.
azimidokht/recipe-recom Text Instruction tuning for nutritional filters, quick meals, and diet queries.
jojogo9/Food_Recipes Tabular Text Broad culinary knowledge and ingredient-to-step associations.
SunnyAgarwal4274/Food_Ingredients Vision Raw produce and ingredient classification images.
snappr/idefics-food-ai Vision High-quality plated dishes natively formatted for multimodal AI.

πŸ“‹ Data Structure

Every sample adheres to the standard Unsloth/Gemma conversational format:

{
  "messages": [
    {
      "role": "user",
      "content": [
        {"type": "image"}, // Present for visual queries, omitted for text-only
        {"type": "text", "text": "What food is this, and can you give me a recipe for it?"}
      ]
    },
    {
      "role": "assistant",
      "content": [
        {"type": "text", "text": "This looks like pizza! Here is a quick way to prepare it: ..."}
      ]
    }
  ],
  "image": "<PIL.Image null or>"
}

πŸ› οΈ Usage with Unsloth / Hugging Face

from datasets import load_dataset

dataset = load_dataset("alst10/gemma4-multimodal-recipe-dataset", split="train") print(dataset[0])

Downloads last month
79