About TokenCost

TokenCost (llmapicosts.com) is a free tool that helps developers and teams understand what large language model APIs actually cost. Per-token pricing is hard to compare across providers — every provider quotes it slightly differently, prices move without notice, and cached-input rates are easy to miss entirely. TokenCost normalizes every model to USD per million tokens and turns those rates into a real monthly dollar figure, so comparison is finally apples-to-apples.

What you can do here

• Pricing — browse every current model’s input, output, and cached-input rate on one honest scale, filterable by provider, release year, and open- vs closed-weight.
• Calculator — enter your expected traffic and project your monthly bill across every model before you ship.
• Compare — put models head to head on price, context window, and a shared cost example.
• Cheapest models — a ranked shortlist by workload, so you don’t have to sort the full table yourself.
• Token counter — paste actual text to see its token count and price it against every model, instead of guessing.
• Benchmarks — see capability against cost across eleven leaderboards spanning coding, reasoning, and math, so you pick the best value, not just the top score.
• Price trends — a running, dated log of real price cuts, increases, and new or retired models, not just a snapshot.

Why it exists

Choosing an LLM shouldn’t require a spreadsheet and ten open tabs. TokenCost exists to make that decision fast, transparent, and honest — no sign-up, no sales pitch, just the numbers.

Sourcing and standards

Every number on this site traces back to a live, automated pipeline rather than a hand-typed spreadsheet — the full technical breakdown of exactly where each figure comes from and how it’s processed is on our methodology page. For how we handle independence, corrections, and advertising, see our editorial policy.

Get in touch

Spotted an incorrect price or have a suggestion? We’d love to hear it — head to our Contact page.