Show HN: Free text classification for your n8n workflows (jeffyclassify.com)

A new community node called Jeffy has been introduced for n8n workflows, providing a reliable alternative for text classification without relying on large language models (LLMs). Jeffy utilizes logistic regression on frozen embeddings, delivering consistent outputs based on input while overcoming the parsing errors and variability commonly found in n8n's built-in Text Classifier that employs LLMs. The Jeffy node offers access to 70 pretrained classifiers for various banking intents, ensuring efficient routing of customer messages—such as directing account closure inquiries to the retention team. This development is significant for the AI/ML community as it highlights a shift towards deterministic models for specific tasks, enhancing reliability and maintaining user privacy by enabling self-hosting options. Jeffy’s functionality is not only faster with lower latency—achieving around 15ms for local processing—but also eliminates the need for API keys and associated costs. With its straightforward setup and real-time classification capabilities, Jeffy provides a valuable tool for automating responses in workflows, ultimately improving the reliability of text classification processes within n8n environments.