Token (LLM Context)
In the context of large language models, a token is the basic unit of text the model processes — roughly a word fragment, word, or punctuation mark — used to measure both input length and output length, with practical limits imposed by the model's context window.
Last reviewed: 2026/05/19
Definition
Why It Matters for Lawyers
How AI Tools Handle It
Frequently Asked Questions
- Q1: How can I tell if a document is too long for a legal AI tool to process accurately?
- Most tools have disclosed maximum input lengths. As a practical test, include a specific fact early in a long document and then ask the tool about it near the end of the session. If the tool cannot accurately recall or reference the early information, it may be operating at or near its context limit. For critical matters, limit document-analysis sessions to within the tool's specified capacity.
- Q2: Does the context window fill up with conversation history as well as documents?
- Yes. In a multi-turn AI conversation, each previous exchange consumes context window tokens. A tool that has processed a 50,000-token contract and then engaged in 20 exchanges has used substantially more of its context window than a fresh session analyzing only the contract. Some tools manage this by summarizing or compressing prior context; others reset the context after a limit is reached.
- Q3: Are tokens the same across all LLM providers?
- No. Different models use different tokenization schemes, so the same text may convert to different token counts across providers. As a rough guide, 1,000 tokens corresponds to approximately 750 English words. Legal text, which tends to have longer words and complex sentence structures, typically tokenizes at a somewhat higher ratio than general prose. --- *Last reviewed: 2026-05-19 by LawyerAI Editorial Team.*
Related Concepts
Context Window
The context window is the maximum amount of text — measured in tokens — that a large language model can process at one time, determining how much document content, conversation history, and instructions the model can consider when generating a response.
Tech / ModelLLM (Large Language Model)
A large language model (LLM) is an AI system trained on large volumes of text data to predict and generate human-like text; it serves as the core engine underlying most legal AI tools for research, drafting, and document analysis.
Tech / ModelInference
In AI, inference is the process of running a trained model to generate outputs from new inputs — as distinct from training, which creates the model. Every time a lawyer submits a query to a legal AI tool, inference occurs.
Related Tools
- Harvey AI
The most expensive legal AI in the market — Am Law 100 firms only.
- CoCounsel
Thomson Reuters' GPT-backed research and drafting with Westlaw integration.
- Westlaw Precision AI
AI-powered legal research with citation-validated answers from Westlaw.
- Lexis+ AI
Conversational legal research with real-time Shepard's citation validation.
- Spellbook
AI contract drafting and review inside Microsoft Word for transactional lawyers.
Related Reading
Last reviewed: 2026/05/19. Definitions are written by the LawyerAI Editorial team. We do not accept affiliate commissions; Featured placement is clearly labeled and does not influence editorial content.