About LLMCost
Live LLM API pricing, compared. Know what every model costs before you call it.
What this site does
Every day, an automated pipeline pulls the full model catalog and pricing from OpenRouter's public API (458+ models), snapshots it into version control, rebuilds every page as static HTML, and deploys. Each price on this site carries:
- a verification date (the build date shown next to the price)
- a source link back to the provider's own listing
- input, output, and cache-read prices per 1M tokens
How we're different
- Scenario-based cost estimates. Tables alone don't answer "what will my app cost?" The calculator and scenario pages rank models by real workload shapes (chatbot, coding agent, RAG…).
- Savings recommendations. The Bill Shredder analyzes your usage CSV entirely in your browser and suggests cheaper models with context-window guardrails.
- Agent-native. AI assistants quote stale training-data prices all the time.
/llms.txtand our JSON API give agents a fresh, citable source. - No paywall, no tracking pixels beyond privacy-friendly analytics. Cloudflare Web Analytics is cookieless and does not fingerprint visitors.
Data sources & accuracy
Prices reflect OpenRouter list prices at build time. Provider first-party pricing can differ (volume discounts, batch tiers, prompt-caching discounts). Always confirm on the provider's official pricing page before making budget decisions — every model page links to its source. Quality scores on the Value page come from the Artificial Analysis Intelligence Index, manually verified per model.
Contact
Corrections, data issues, or partnership inquiries: hello@llm-cost.io. See also our Privacy Policy and Terms of Service.