How is the token count calculated?

Legal English tokenizes at roughly 4 characters per token, but dense defined terms and numbering push it higher. The tool uses a legal-aware ratio and shows the estimate, which is typically within a few percent of a tokenizer-exact count.

Is my document uploaded anywhere?

No. Counting and cost estimation run entirely in your browser. Nothing is sent to us or any provider, which matters for privileged or confidential material.

When should I chunk instead of one long call?

Chunk when the document approaches or exceeds the model's window, when you need clause-level citations, or when long-context recall is unreliable for your task. The tool flags chunking once the input plus expected output nears the window.

Why do output tokens vary by task?

A one-paragraph summary emits few tokens, while clause-by-clause extraction or a redline can emit as many tokens as the input. The tool scales expected output by task type so the cost reflects the real workload.

Should I trust the model with legal analysis?

LLMs are useful for triage, summarization, and surfacing clauses, but they hallucinate and miss nuance. Treat outputs as a first pass for a qualified reviewer, never as legal advice or a substitute for counsel.

Legal Document Token Cost Estimator

Legal document token cost estimator

Running contracts and filings through an LLM is increasingly common for summarization, clause extraction, and risk triage — but per-document cost adds up fast, and long legal text often brushes against model limits. This estimator counts the tokens in your document, projects the output tokens your task will generate, and prices the analysis across common models so you can budget and choose between a single long-context call and a chunked pipeline.

How it works

You paste the document and pick an analysis task. The tool estimates input tokens using a legal-aware characters-per-token ratio, then projects output tokens by scaling against the task: a summary emits a fraction of the input, while full clause extraction or a redline can match it. It multiplies input and output token counts by each model’s input and output prices to give a per-analysis cost, and checks whether input plus expected output fits a single call or needs chunking. All of this happens locally — the text never leaves your browser, which is essential for confidential material.

Tips and notes

Keep it local for privilege. Counting here is offline; only your eventual real API call sends data to a provider.
Match the task. Picking “Q&A” when you actually need clause extraction will underestimate output cost.
Chunk for citations. Clause-level extraction is more reliable and auditable when each section is its own call.
Not legal advice. Use AI output as triage for a qualified reviewer, never as a final legal opinion.