Explore AI models
Compare with confidence

Every model on this site is priced the same way: USD per one million tokens, shown separately for input and output. Providers don't all quote prices in that same unit on their own pages — some use per-thousand-token pricing, some show monthly tiers — so every price used here is converted to the same per-million-token basis before it reaches a calculator. That makes models comparable to each other, but it does not change what the provider actually charges.

• Budget-friendly — capable, low-cost models that handle most structured support FAQs and high-volume tasks. Often the best value when the task doesn't need frontier reasoning.
• Balanced — mid-tier models that trade a higher price for stronger reasoning and longer, more nuanced answers. A common default for mixed-complexity workloads.
• Frontier — flagship models for the hardest reasoning. They cost the most per token, so they pay off when quality genuinely depends on them, and can be wasteful for simple, repetitive tasks.

AI provider prices change more often than most reference pages get updated. The prices behind every calculator on this site were manually checked against each provider's own pricing page as of 2026-09-08 — see the pricing reference for the verification date and caveats. Before committing a budget, re-check the number against your provider's current, official pricing page.

• Estimates are for planning only — they are not a billing guarantee from any provider.
• Prices shown are the standard per-token rate; some models apply a higher rate above a long-context threshold (noted on the model's own card where it applies).
• Provider pricing changes over time — always re-check the official pricing page before committing a budget.