Independent web application
llm-spend
Putting AI costs into context.
A pricing reference and comparison tool for LLM APIs, with provider sources, cache rates and workload-based calculations.
There is more than one price.
Input tokens, output tokens, cached input and deployment tiers can all affect an API bill. llm-spend puts these distinctions alongside provider references so comparisons have context.
From a table to a decision.
The application includes a workload calculator, comparison views, shortlists and a budget planner. Prices can be viewed in USD and CHF, with source notes that distinguish published rates from derived figures.
Make the sources inspectable.
Pricing records, calculation logic and updates live in the public TypeScript repository. The website links to source information and exposes a freshness view. A quoted API price is a dated reference, not a guarantee of a future bill.
NEXT PROJECT