Builder
$20/mo
1x monthly capacity
For individual builders and everyday agent work.
- Complete public catalog
- Supplier-weighted capacity
- No overage billing
Secure Stripe checkout · Cancel anytime
50% off
GPT-6 vs list
70% less
Claude Fable 5 input
$2
Pay as you go start
1 key
Every frontier model
Generate anything through a single, unified interface. All major models in one place.
Browse catalogReliable models through our distributed routing. Traffic fails over automatically.
View statusKeep costs in check without sacrificing speed. Discounted rates up to 70% below list.
See pricingProtect your organization with fine-grained routing. Your prompts go only where you allow.
Read docsEvery model in one catalog, one key
Create an account to get started. You can set up an org for your team later.
Credits can be used with any model or provider. Top up from $2, or pick a monthly package.
Create an API key and start making requests. Fully OpenAI compatible.
$20/mo
1x monthly capacity
For individual builders and everyday agent work.
Secure Stripe checkout · Cancel anytime
$100/mo
5x Builder capacity
For serious coding sessions and larger projects.
Secure Stripe checkout · Cancel anytime
$200/mo
20x Builder capacity
For launches and sustained high-capacity workflows.
Secure Stripe checkout · Cancel anytime
$350/mo
50x Builder capacity
For teams and always-on agent workloads.
Secure Stripe checkout · Cancel anytime
Prefer variable spend?
Pay as you go from $2. No subscription, no automatic refill.
FAQ
Yes. Xpersona routes to genuine GPT, Claude, Gemini, and Xpersona-original endpoints. The full catalog, benchmark methodology, and observed results are published so you can inspect the record before you buy.
Yes. Monthly packages (Builder, Pro, Studio, Scale) are secure Stripe subscriptions you can cancel at any time. Pay-as-you-go has no subscription, no automatic refill, and no lock-in.
Each package states its monthly capacity up front. Supplier-weighted usage means premium models draw the allowance down faster than lighter ones. At the limit, requests pause instead of generating surprise overage charges.
Yes. Top up from $2 at any time. GPT-6 PAYG runs at 50% below official OpenAI Standard API pricing, with cached input and output billed per the same discounted rates. There is no subscription and no automatic refill.
Anything OpenAI-compatible: OpenCode, popular chat clients, CLIs, and your own code through the same /v1 endpoints. Set one base URL and one API key and every tool keeps working.
You can start with $2 of prepaid credits to evaluate the full catalog. Some packages also offer trials — check the pricing page and your dashboard for current trial eligibility.
A copy-pasteable path from an empty project to a multi-model request, including model discovery, error handling, and a production checklist.
The right coding model depends on repository shape, task type, tool loop, latency budget, and how you judge a correct change.
A router should make a measurable decision, not conceal an unpredictable chain of model fallbacks.