Short limits that come back fast
Each 5-hour window has its own allowance. Use it up in a long session and it is back five hours after you started.
An AiCheap Plan gives your coding agent Opus, Sonnet, Fable, GPT and Gemini through one dedicated API key. Usage refills every 5 hours and every week. Prefer no commitment? The prepaid wallet stays pay-as-you-go.
AiCheap Plans
Every plan includes the same models and a dedicated API key. Bigger plans give more usage at a better rate. Pick a model to see what each plan covers.
* Token estimates assume a coding-agent workload of 95% cache reads and 5% cache writes, with the whole allowance spent on one model. “At list price” is the same usage billed at the vendor's official API prices. Usage refills every 5 hours and every week.
Each 5-hour window has its own allowance. Use it up in a long session and it is back five hours after you started.
The week caps total use and resets on the day you subscribed, so one heavy week never ends your plan.
Optional and off by default. Turn it on and requests continue from your prepaid wallet at pay-as-you-go prices until the plan refills.
Set your weekly usage. We compare the vendor's list price, the AiCheap wallet and the smallest plan that fits.
Console & wallet
Create an account, fund a prepaid wallet, and use chat to compare models side by side. When you are ready, issue API keys for the same balance and billing profile—no separate procurement step.
Chat and API usage draw from one wallet. Catalog prices and balances are shown in IDR; fund via QRIS, VA transfer, bank transfer, PayPal, or cryptocurrency.
Threads, model switching, and spend visibility—so teams can pilot models before committing spend.
New models appear in the live catalog below with descriptions and per-million IDR pricing for comparison.
Top up as needed, or take an AiCheap Plan from Rp 178.000 every 4 weeks when usage is steady. See plans →
API & integration
Point Cherry Studio, CC Switch, internal services, or your product at AiCheap’s base URL. Change model in the request body to switch vendors without rewriting client code.
Typical integration
BASE_URL=https://api.aicheap.ai/v1
OPENAI_API_KEY=sk-… # from your AiCheap dashboard
POST /v1/chat/completions
# Claude, Gemini, embeddings, images — see live catalog
SDKs, agents, and IDE plugins that expect an OpenAI base URL typically work after you swap host and key. Each model lists supported endpoint types.
Usage logs include tokens and latency per call—useful for cost control and model A/B tests.
Traffic is routed across channels with fallbacks to reduce impact from upstream outages.
The rankings below reflect real routed volume on AiCheap—not a static marketing list—so you can see what teams actually run.
Rankings and model prices on this page are loaded from the same APIs that power the console—refreshed every 90 seconds. Compare cost and adoption before you pick a default model.
Loading…
Loading models…
Top up with QRIS, VA transfer, bank transfer, PayPal, or cryptocurrency—one prepaid balance for chat and API usage.