Hey Product Hunt 👋 I'm Soufiane, the maker of LLM Cost Lab. Pricing an AI feature usually means ten provider tabs, different units, and a spreadsheet that is out of date by next month. And the real bill rarely matches the headline per-token price, because caching, batch discounts and output-heavy prompts change the math. So I built one place that answers "what will this actually cost me?": 📊 88 models from 18 providers in one sortable table, input, output and cached rates per 1M tokens 🧮 17 free calculators: API cost, cost per request, token counter, RAG pipeline, AI agent cost, prompt caching and batch savings, fine-tuning, embeddings, GPU rental and more 📈 A price index for the whole market: among the models we track, input rates span $0.03 to $30.00 per 1M tokens, a 1000x spread The part I care most about: every rate is read from the provider's own pricing page and carries the date it was checked. Retired models get flagged instead of quietly left in. It's free, no signup, nothing to install. If you run an API, you can also list it (one-time, clearly labelled, and it never changes a price or a ranking). I'd love your feedback: which model or calculator is missing for your use case? 🙏

LLM Cost Lab

Compare LLM API prices and calculate your real AI bill

Compare LLM API prices and calculate your real AI bill