PromptSpend
PromptSpend is a free, browser-based cost calculator for large language model APIs. A user pastes a prompt or describes an expected workload, and the tool shows estimated bills for the same task side by side across a catalog of roughly 78 models — including GPT, Claude, Gemini and other provider families. The result is a per-provider and per-model comparison of what an AI feature would cost to run at the given volume. The project's stated design premise is that pricing comparison tools decay quickly because vendors change their rates without warning. Rather than ship a static price table, PromptSpend treats the pricing pipeline itself as the product: the underlying model catalog is re-checked every morning so that comparisons reflect current published rates instead of stale snapshots. The site is described as free and open source, with no account creation and no analytics tracking. It targets developers, ML engineers and product people who need an early order-of-magnitude cost estimate before committing to a model or architecture. Because pricing is verified on a daily cadence, the tool is positioned as a lightweight pre-build check — "know the tab before you build" — rather than a full observability or spend-management platform.
- Category
- Prompt Engineering Tools
- Pricing
- Unknown
Problem solved
LLM API pricing changes frequently, so static cost comparison tables and blog-post estimates go stale quickly and quietly mislead teams budgeting an AI feature. PromptSpend addresses this by maintaining a model price catalog that re-checks provider rates every morning, so cost comparisons across dozens of models reflect up-to-date published pricing rather than outdated snapshots.
Use cases
- Estimate the monthly API cost of a planned AI feature before writing code
- Compare per-task cost across GPT, Claude, Gemini and other model families
- Sanity-check a model choice when a cheaper alternative may meet requirements
- Budget token spend for a prototype or internal tool by expected request volume
- Reference current published model prices without hunting through each vendor's pricing page
Audience
- Developers
- ML Engineers
- Product Managers
- Startups
Pros
- Free to use with no account or signup required
- Prices re-checked daily, reducing staleness risk
- Compares a broad catalog of roughly 78 models side by side
- No analytics tracking reported
- Open source, so the pricing pipeline can be inspected
Cons
- Accuracy still depends on provider pricing pages being parsed correctly
- Focus is pre-build estimation, not runtime spend monitoring
- Limited information available about supported deployment or self-hosting
- No public GitHub repository linked from the homepage at the time of review
- Pricing for the tool itself is unspecified
Official website: https://promptspend.com/
Canonical LambdaBase page: https://www.lambdabase.com/tools/promptspend