What is a token in LLM APIs?
A plain-language explanation of tokens and why they drive AI API bills.
Read articleLearning Center
Learn how AI pricing and token costs actually work
Guides, tutorials, and concepts that connect understanding to calculation—so you can estimate spend, compare models, and optimize with confidence.
Topics
What you can learn here
Start with a lane—guides for planning, tutorials for calculators, concepts for the foundations.
Start here
Foundational reads
Core explainers that unlock clearer calculator and comparison decisions.
How to estimate LLM API costs before you ship
A practical framework for turning traffic assumptions into monthly OpenAI, Anthropic, and Google API spend.
Read articleHow much do 1 million tokens cost?
A plain explanation of per-1M input and output prices—and why your real 1M-token bill depends on the mix.
Read articleInput vs output tokens — a simple guide
Why output tokens often dominate the bill, and how to estimate both sides before you call an API.
Read articleLibrary
Latest articles
25 articles covering budgets, routing, and core concepts.
Input vs output tokens — a simple guide
Why output tokens often dominate the bill, and how to estimate both sides before you call an API.
Read articleWhy tool-calling agents cost more than single-shot chat
How growing context, tool-result tokens, and multi-turn loops drive agent API bills—and how to estimate them.
Read articleHow prompt caching reduces LLM API costs
Separate stable prefixes from dynamic tokens, estimate hit rate, and account for cache-write fees when planning spend.
Read articleChoosing models by latency vs API cost
How to set a latency SLO, compare planning p50 latency with list prices, and avoid overpaying for speed you do not need.
Read articleHow to pick the cheapest model without breaking quality
Use task presets, capability floors, and evals so cost routing stays safe in production.
Read articleUnit economics for SaaS AI features
Connect token costs, adoption, and pricing so AI features stay within a healthy COGS band.
Read articleAPI resale margin: cost vs what you charge
How to price AI wrappers and products so provider COGS, payment fees, and overhead still leave healthy margin.
Read articleConvert USD AI API bills to GBP, EUR, AUD, and CAD
How UK, EU, Australian, and Canadian teams should budget OpenAI and Anthropic invoices that arrive in dollars.
Read articleHow to analyze an LLM usage CSV without uploading it
Export model and token columns, match them to list prices in the browser, and spot invoice variance.
Read articleWhen self-hosted LLMs beat API pricing
Compare GPU-hour economics to hosted token prices, and know which assumptions flip the decision.
Read articleWhy bigger context windows cost more
How 8K vs 128K vs 1M prompts change token bills—and when caching or retrieval beats stuffing everything in.
Read articleHow API rate limits shape LLM cost and capacity
RPM, TPM, concurrency, and daily caps decide your real ceiling—often before list price does.
Read articleAllocate AI API budget across team seats
Use role weights and hard caps so a few power users do not consume the whole monthly AI pool.
Read articleHow much do 1 million tokens cost?
A plain explanation of per-1M input and output prices—and why your real 1M-token bill depends on the mix.
Read articleWhy retries inflate AI API cost
How rate limits, timeouts, and partial billing turn a “successful job” budget into extra attempts and spend.
Read articleHow to estimate PDF and document ingestion cost
Turn pages into tokens, account for chunk overlap, and decide when an LLM extract pass is worth the spend.
Read articleDoes streaming change LLM API cost?
Why streamed and buffered responses usually share list prices—and when early cancel or wait economics tip the comparison.
Read articleHow to price A/B and cascade model routing
Estimate LLM spend when you split traffic or escalate hard cases from a cheap primary to a premium model.
Read articleHow to estimate multi-provider failover cost
Price an availability path: primary success rate, failed-attempt billing, and secondary spend during blips or outages.
Read articleHow to estimate LLM-as-judge eval cost
Break eval spend into candidate generation and judge scoring—and see why rubric dimensions multiply the bill.
Read articleHow to estimate synthetic data generation cost
Price generate → critique → filter pipelines and understand why accept rate drives over-generation spend.
Read articleWhen does a RAG re-ranker pay for itself?
LLM rerank adds calls, but feeding top-N instead of top-K into the generator can cut context tokens enough to win overall.
Read articleHow to estimate LLM API costs before you ship
A practical framework for turning traffic assumptions into monthly OpenAI, Anthropic, and Google API spend.
Read articleBatch API vs realtime: when to switch
Learn how to compare batch and realtime pricing for the same token workload with CentsPerToken.
Read articleWhat is a token in LLM APIs?
A plain-language explanation of tokens and why they drive AI API bills.
Read article