Billing & Credits

Pay-as-you-go pricing with no monthly fees. Add credits when you need them.

How pricing works

You pay the model's base cost + 10% markup. That's it. No subscriptions, no minimum spend, no hidden fees.

For example, if a GPT-4o call costs $0.01 on OpenAI, it costs $0.011 through LLMTest. You get cost tracking, fallbacks, and JSON recovery included.

Adding credits

Purchase credits from your dashboard. Available amounts: $5, $10, $25, $50, $200. Credits never expire.

Auto top-up

Enable auto top-up in settings to automatically add credits when your balance drops below a threshold. Requires at least one prior purchase (so we have a payment method on file).

Checking your balance

• Dashboard: Shows current balance, total spend, and per-flow costs
• API: GET /api/account returns credits_usd and total_spent
• MCP: Ask your assistant "What's my LLMTest balance?"

What uses credits

• Proxy calls — each call through /v1/chat/completions is charged at model cost + 10%
• Benchmarks — running benchmarks uses credits for the judge model and challenger inference. Typically costs a few cents per benchmark.