Know what every AI model really costs.

TokenSwarm is a free tool that answers one question developers ask constantly: how much does this model cost per token? It brings live pricing for 300+ large language models — from OpenAI, Anthropic, Google, Meta, Mistral and dozens more — into one place, with a calculator that turns those tiny per-token prices into a real estimate of your spend.

Where the data comes from

Model names, context lengths and input/output token prices are pulled live from the public OpenRouter model catalog, which aggregates pricing across providers. Prices update automatically. We don't set or alter any figure — everything shown is the published rate.

An estimate, not a quote

Provider prices can change at any time, and your real bill depends on exact token counts, caching, and provider-specific rules. Treat TokenSwarm as a fast comparison and planning tool, and always confirm the final price on the provider's own pricing page.