HyperFast is a novel meta-trained hypernetwork designed for instant classification of tabular data in a single forward pass. It generates task-specific neural networks tailored to unseen datasets, eliminating the need for traditional model training. The method is evaluated on a wide range of datasets, including genomic data and OpenML datasets, demonstrating highly competitive results compared to other tabular data neural networks, traditional ML methods, AutoML systems, and boosting machines. HyperFast shows significant speed improvements while maintaining robust adaptability across various classification tasks with minimal fine-tuning. The approach is particularly useful for applications requiring rapid model deployment, such as healthcare and data streaming. The code and trained HyperFast models are available at https://github.com/AI-sandbox/HyperFast.HyperFast is a novel meta-trained hypernetwork designed for instant classification of tabular data in a single forward pass. It generates task-specific neural networks tailored to unseen datasets, eliminating the need for traditional model training. The method is evaluated on a wide range of datasets, including genomic data and OpenML datasets, demonstrating highly competitive results compared to other tabular data neural networks, traditional ML methods, AutoML systems, and boosting machines. HyperFast shows significant speed improvements while maintaining robust adaptability across various classification tasks with minimal fine-tuning. The approach is particularly useful for applications requiring rapid model deployment, such as healthcare and data streaming. The code and trained HyperFast models are available at https://github.com/AI-sandbox/HyperFast.