Coding plans
Frontier tokens and theta requests, every month. Prices shown for India — international checkout in USD. Cancel anytime.
basic
- 20M frontier input / mo
- 5M frontier output / mo
- 400 theta requests / 5h (up to ~58k / mo)
- GLM-5.3, Qwen-3.8 & theta endpoints
- OpenAI + Anthropic compatible
advanced
- 500M frontier input / mo
- 100M frontier output / mo
- 8000 theta requests / 5h (up to ~1000k / mo)
- GLM-5.3, Qwen-3.8 & theta endpoints
- OpenAI + Anthropic compatible
How we sustain these prices
Three engineering choices, all disclosed: prompt-cache engineering (stable prefixes so cache hits stay high), context management (your history stays lean), and smart routing (heavy reasoning gets the full model; execution runs on the Flash tier). No hidden model substitution — you always know which endpoint you called.
We may use API traffic to train our own models (our theta research). You can opt out at any time from your dashboard — one toggle, no email required.
What would this cost at direct API rates?
Compare our plan price against buying the same tokens directly — at list rates, no caching tricks.
theta — metered display rates
| Model | Input / M tokens | Cache hit / M | Output / M |
|---|---|---|---|
theta | $0.2 | $0.04 | $0.4 |
theta is request-metered in plans (see cards above). These display rates power the equivalent-cost view in your dashboard and the calculator.