Glossary term
Glossary term
Foundations
Maximum amount of tokens a model can consider in one request.
The number of tokens a model can process in a given prompt. The larger the context window, the more information the model can use to provide coherent and consistent responses to the prompt.
A financial analyst submits a 90-page earnings report (~70k tokens) to Claude 3.5 to extract all forward-guidance statements and quantitative projections in one pass - impossible with a 4k-token model.
Google Gemini 1.5 Pro's 1M-token context window lets a video production company embed an entire hour-long documentary transcript and ask the model to identify all scenes mentioning a specific brand.
A law firm uses a 200k-token window to compare two versions of a 150-page acquisition agreement, instructing the model to flag every clause that changed and summarise the legal risk delta.
Created for this library
A legal-tech vendor selects an LLM with a 200,000-token context window so it can review entire merger agreements in a single pass.
A coding assistant team uses a long-context LLM so the model can see the whole open repository, not just the current file, when suggesting changes.
A research assistant tool truncates documents to fit the context window and adds RAG so the most relevant passages survive truncation.
Definition source: Google for Developers Machine Learning Glossary | Creative Commons Attribution 4.0 License