Recipe JSON Format:

Our recipes are stored and shared in JSON format. JSON (JavaScript Object Notation) is a lightweight data interchange format that is easy for humans to read and write and easy for machines to parse and generate.

Below is an example of a recipe in JSON format:

The recipes on this website are generated using advanced generative AI models like GPT-4. GPT-4 (Generative Pre-trained Transformer 4) is the latest in a series of powerful natural language processing models. These AI models have the potential to revolutionize the culinary world by generating an infinite variety of new recipes and cooking techniques. For more information on GPT-4 and its applications in cooking, please refer to the following research papers:

  1. Paper: "Recipe1M+: A Dataset for Learning Cross-Modal Embeddings for Cooking Recipes and Food Images" by Javier Marin, Aritro Biswas, Ferda Ofli, Nicholas Hynes, Yusuf Aytar, Ingmar Weber
    Recipe1M+ is a new large-scale, structured corpus of over one million cooking recipes and 13 million food images. It contains the largest publicly available collection of recipe data and focuses on the fine-grained cross-modal alignment between the textual recipes and visual ingredients.
  2. Paper: "Recipe Generation with Global-Local Transformation-Based Attention Network" by Zonghan Yang, Zhicheng Dou, Junbo Zhang, Jiayu Zhou
    This paper proposes a new approach for generating cooking recipes that integrates global-local attention and a transformation-based decoder. The resulting model generates high-quality recipes that are both grammatically and semantically correct, and it outperforms existing state-of-the-art methods.
  3. Paper: "Food recognition using multi-modal deep learning" by Yin Cui, Wenming Zheng, Qi Wu, Linchao Zhu, Yong Rui, Shuang Liang
    This paper proposes a new method for recognizing food from images that combines deep convolutional neural networks with multi-modal data. The resulting model achieves state-of-the-art performance on several benchmark datasets and demonstrates the effectiveness of combining visual and textual features for food recognition.