NER: Gemini vs Spacy vs Compromise

TLDR For NER, if accuracy is critical, go with an LLM — even an old one like gemma-3-27b-it will outperform tools or small models trained for this task. But by using an LLM you are exposing your data, making an HTTP request, and most likely incurring a cost. If accuracy is not critical and you want to stay in Javascript, compromise is a good package for NER. If you want an even better package and it's OK not using Javascript, then try Spacy. Intro Thanks to AI it feels like we are entering into Web 4 — now it's not just about having a static website (Web 1), or letting the user save their data (Web 2), or decentralization (Web 3, although the majority of companies still own our data); but about adding "AI" to your application. What does that mean? Are all applications supposed to be like ChatGPT? While I don't think adding an LLM just for the sake of it is beneficial, I do think that playing with LLMs helps to learn about them, giving us an understanding of how, when, and why to use th