OpenLuxSimple, transparent pricing
Every plan includes OpenAI-compatible endpoints, the full model catalog and real-time usage details. Upgrade or switch at any time.
Choose the plan that fits
Every plan includes OpenAI-compatible endpoints, the full model catalog and real-time usage details. Upgrade or switch at any time.
Starter
Includes
- 2M token-equivalent credit
- All text models
- OpenAI-compatible endpoints
- Standard concurrency
- Usage details in the console
- Ticket support
Pro
Everything in Starter, plus
- 20M token-equivalent credit
- All models: text, image, audio & video
- Higher concurrency and priority routing
- Request logs and spend analytics
- Priority ticket support
- Multiple API keys
Team
Everything in Pro, plus
- 80M token-equivalent credit
- Member and permission management
- IP allowlists and key scoping
- Named support engineer
- Invoicing and reconciliation
- Enterprise SLA options
Prices are billed in USD and metered on actual usage. Per-model rates are published in the model catalog. Need higher volume or an enterprise agreement? Contact us.
Forecast a workload with the LLM API cost calculator, or compare coding-agent billing in the agent cost comparison.
Estimate spend before you scale
Model a workload, compare coding-agent billing and understand where tokens, retries and repeated context increase the invoice.
LLM API cost calculator
Estimate sessions, tokens, retries and team spend.
AI cost optimization
Reduce context, retries and model-tier waste.
Coding agent comparison
Compare client plans, API usage and cost per task.
Claude Code cost
Budget long-context coding sessions and retries.
Codex CLI cost
Forecast terminal-agent tasks and monthly team spend.
Cursor API cost
Separate the editor plan from custom-endpoint usage.
Flexible options for enterprises, resellers and model providers
Dedicated routes, custom capacity, SLA coverage and compliance support — with contracts, bank transfer and invoicing.
Pay as you go or a monthly plan?
Choose pay as you go when usage is spiky, experimental or hard to forecast. You are billed per token or per request at the published rate for the model you call, balances do not expire, and there is no minimum commitment. This is the cheaper route until your monthly spend approaches the price of a plan.
Choose a monthly or yearly plan when spend is predictable and already close to the plan price. A plan adds a token-equivalent credit allowance, higher concurrency and priority routing, and yearly billing gives two months free compared with paying monthly. Plans can be upgraded, downgraded or cancelled from the console at any time.
Whichever route you take, the per-model rates in the model catalogue are the same, and requests are metered on the tokens or calls you actually use. If you need committed capacity, dedicated routes or an SLA, talk to us about an enterprise agreement.
What every plan includes
- One OpenAI-compatible endpoint and one API key for all models
- The full catalogue — text, image, audio, video and embeddings
- Per-request logs with model, latency, token count and cost
- Scoped keys with spend limits and model permissions
- No separate provider accounts, invoices or minimum spend
Billing questions
Can I cancel a subscription at any time?
Yes. Cancel from the console and the plan simply stops renewing at the end of the current period; usage already consumed is not refunded.
What happens when my quota runs out?
Top up pay-as-you-go credit from the console or move to a larger plan. Credit balances do not expire.
Which payment methods do you support?
Card payments through Stripe, with invoices available. Enterprise customers can request bank transfer and a contract.
Do you offer enterprise plans?
Yes — dedicated routes, custom capacity, SLAs and compliance support. Get in touch and we will scope it with you.