Model endpoints & pricing

Every model, one transparent price.

Pay per token, in and out, with no idle GPU charges and no infrastructure to manage. Base flavor for cost-efficient throughput, Fast flavor when latency is what matters. Get an instant estimate below, or see the full catalog.

Billing and endpoints available for Southeast Asia accounts only — Singapore, Indonesia, Malaysia, Thailand, Vietnam, Philippines

Get an instant quote

Estimate your monthly cost for any model on Hi Liberty. This is a ballpark based on list price — volume discounts below can bring it down further.

Estimated monthly cost
$0
—
Input cost$0
Output cost$0
Rate (in / out per 1M)$0 / $0

List pricing shown. Volume discounts apply automatically at checkout above 1B tokens/month — see tiers below. Need a dedicated endpoint instead? Talk to sales for a fixed quote.

Full catalog

List pricing, per 1M tokens

Model Context Base ($/1M in · out) Fast ($/1M in · out)
Volume discounts

The more you serve, the less you pay per token.

Starter

List price

0 – 1B tokens / month
  • Shared endpoints, all 70+ models
  • Standard rate limits
  • Community support
Growth

−15%

1B – 20B tokens / month
  • Automatic discount, no negotiation
  • Priority rate limits
  • Email + chat support
Enterprise

Custom

20B+ tokens / month, or dedicated
  • Dedicated endpoints, reserved capacity
  • 99.95% SLA, regional deployment in SG / ID / TH
  • Dedicated support channel

Need a fixed quote for a dedicated endpoint?

Tell us your expected volume and latency targets — our Southeast Asia solutions team will size an endpoint and send back per-token pricing within one business day.

Talk to sales Start free instead