Multi-Model Token Cost Calculator

Token counts here use a rough character-based approximation, not a real tokenizer, so treat this as a planning estimate rather than an exact figure. Prices are standard-tier list prices per 1M tokens (input / output), checked on 12 November 2026 against Anthropic's, OpenAI's and Google's own pricing pages. They change often, and each provider sells more models than the eleven shown. Prompts above a model's long-context threshold cost more (over 100K tokens for Claude Haiku 5.5, over 200K for gemini-3.1-pro-preview, over 272K for the gpt-6 models), gemini-4-argon is priced at twice the rate shown from 1 December 2026, and gemini-3.8-flash from 1 January 2027. Check each provider's current pricing page before finalizing a budget. This tool doesn't account for prompt caching, batch discounts, or volume pricing, all of which can meaningfully reduce real-world cost.