← jwatte.com

LLM Retrieval Cost Estimator

Every page you want an LLM to retrieve costs tokens. Tokens cost money at scale. Different tokenizers (OpenAI cl100k, Anthropic, Google) count the same text differently. This tool estimates tokens-per-page per major tokenizer, multiplies by current per-1K pricing, and suggests chunking strategy for RAG ingestion.

Context and background

Read the story behind this tool: The token-efficiency audit every content team ignores until the bill arrives →

Input

Or paste text directly: