Pricing

LLM API has two pricing tracks. Pick the one that matches how you use the API — you can hold keys on both at once.

AI for Coding — flat monthly subscription

For coding assistants and IDE tools. Each subscription key gets its own monthly (or yearly) budget, refreshed on three overlapping windows so a burst doesn't exhaust your whole month:

Tier	Every 5 hours	Every day	Per month
Lite	1.5M credits	4M credits	20M credits
Pro	5M credits	13M credits	80M credits
Max	11M credits	26M credits	130M credits

Yearly billing = 10 months charged, 2 months free.
New orgs get a one-time 50% off the first month on any tier (monthly billing only).
Exact current $/month for each tier is set per plan and can change — see the live plan picker in the dashboard for today's price.

See Provider fallback / Intelligent routing for what a subscription key can call.

AI for Production — pay-as-you-go with volume discounts

For user-facing features and production workloads. You pay for what you use (credits or your own provider keys via Custom Providers), with the discount rate increasing at higher monthly spend:

Tier	Monthly spend	Discount
Build	$0 – $999	5%
Ship	$1,000 – $9,999	10%
Scale	$10,000 – $99,999	15%
Enterprise	$100,000+	up to 50% (custom)

New orgs get a one-time deposit bonus (up to $1,000, matched at +100%) on their first top-up.

What's included on every plan

Both tracks include: unlimited API keys and IAM rules, auto routing and fallback across providers, usage & cost analytics, zero data retention by default, and prompt management.

Storage & extras

Some features bill separately from the base subscription/PAYG rate:

Full data retention (storing complete request/response payloads instead of metadata-only) — $0.01 per 1M tokens. See Data Retention.
Overage / credit overflow on subscription keys, and provider-direct costs when using your own keys — see Cost Breakdown.