Introduction
Use Claude, GPT, Gemini, DeepSeek and more for less than list price, with one API key, one prepaid balance and per-token billing.
Any model. Pay less per prompt.
PromptsForLess gives you Claude, GPT, Gemini, DeepSeek, Qwen, GLM and other models for less than list price. One API key works for every model, and one prepaid balance pays for all of them. You pay per token, with no subscription and no minimum.
We match your first $5
Add $5 to your balance and we add $5 more. Create your account and API key in the dashboard.
Your first request
Connect your tools
Explore models
Two familiar APIs
| API | Base URL | Use it with |
|---|---|---|
| OpenAI-compatible | https://api.promptsforless.com/v1 | OpenAI SDKs, Vercel AI SDK, LangChain, Codex CLI, Cursor, OpenCode, Zed, n8n, Open WebUI and most other tools |
| Anthropic-compatible | https://api.promptsforless.com | Claude Code and Anthropic SDKs |
The OpenAI-compatible API supports Chat Completions, streaming, tool calls and structured outputs. The Anthropic-compatible API serves the Messages endpoint that Claude Code and the Anthropic SDKs expect.
Both APIs use the same key and the same balance. Model IDs are our own public IDs, such as claude-sonnet-5-5, gpt-6-luna and deepseek-v4.1-flash. Find every ID in the model catalog.
How requests run
- Close to you. Requests enter through edge gateways near the caller, so routing adds little overhead.
- On the model you name. Every request runs on the exact model you ask for, from its maker. We never swap in a substitute.
- Private by default. We keep usage and billing metadata, such as token counts and timestamps. We never store your prompts or responses.
- Billed only on success. Failed requests are not billed.
How token pricing works
Input tokens cover your prompt and any conversation history you send. Output tokens cover the model's answer, including any billable reasoning tokens. A long conversation costs more per turn than a fresh one, even when your latest message is short.
Each model has its own input and output price. Compare them on the home page price board or on each model's page in the catalog.