Pricing

Prepaid balance. Pay for what you use — never a negative bill.

Billed by model price and actual usage; discounted accounts pay the discounted rate. Every call is itemized.

Standard

1.00x

Billed by model price and actual usage.

Get started
Recommended

Pro

Configurable

Custom discounts, per-key budgets and billing export.

Get started

Enterprise

Let's talk

Dedicated pricing and priority support for high-volume teams.

Get started
Encrypted, secure paymentPay by real usage · no lock-inDirect from 8 leading providers

Top up balance

Calls stop when balance runs out — no overdraft.

Temporary hold

Large requests are held at an estimated cap to prevent over-spend.

Actual settlement

Settled by real usage when the call completes (cache hits cost less), itemized in your bill.

Traceable log

Key, model, status, and cost details land in your transaction log.

Model price reference

Prices are in USD per million tokens, read live from the platform pricing config.

glm-5.2Limited-time promo -45%Z.ai
Input:$1.4 / M$0.77 / MOutput:$4.4 / M$2.42 / MCache:Hit $0.26 / M
Available
deepseek-v4-proLimited-time promo -41%DeepSeek
Input:$2.4 / M$1.42 / MOutput:$4.8 / M$2.83 / MCache:Hit $0.14 / M
Available
glm-5.1Limited-time promo -31%Z.ai
Input:$1.4 / M$0.97 / MOutput:$4.4 / M$3.04 / MCache:Hit $0.26 / M
Available
kimi-k2.7-codeLimited-time promo -30%Moonshot AI
Input:$0.95 / M$0.66 / MOutput:$4 / M$2.8 / MCache:Hit $0.19 / M
Available
kimi-k2.6Limited-time promo -10%Moonshot AI
Input:$0.95 / M$0.85 / MOutput:$4 / M$3.6 / MCache:Hit $0.16 / M
Available
kimi-k2.5Limited-time promo -10%Moonshot AI
Input:$0.6 / M$0.54 / MOutput:$3 / M$2.7 / MCache:Hit $0.1 / M
Available
MiniMax-M2.5Limited-time promo -10%MiniMax
Input:$0.3 / M$0.27 / MOutput:$1.15 / M$1.03 / MCache:Hit $0.06 / M
Available
deepseek-v4-flashLimited-time promo -35%DeepSeek
Input:$0.2 / M$0.13 / MOutput:$0.4 / M$0.26 / MCache:Hit $0.03 / M
Available
qwen3.7-plusLimited-time promo -25%Qwen
Input:$0.4 / M$0.3 / MOutput:$1.6 / M$1.2 / MCache:Hit $0.08 / M
Available
qwen3.7-maxLimited-time promo -41%Qwen
Input:$2.5 / M$1.47 / MOutput:$7.5 / M$4.42 / MCache:Hit $0.5 / M
Available

See full pricing at the Model Hub.

FAQ

Common questions about billing and top-ups.

Top up directly in the console: open the Balance & billing page, pick an amount, and pay with international cards and other online methods — all payments settle in USD. Credit arrives automatically within seconds. For bank transfers or large top-ups, contact [email protected].

API calls return an error when the balance is insufficient — there is no overdraft. You can check your balance and spending in real time on the Balance & Billing page in the console.

Topped-up balance is generally non-refundable. Abnormal charges caused by platform faults will be verified and refunded by our support team. See the billing guide for details.

For large requests, the system first holds the estimated maximum from your balance to prevent over-spend. Once the request completes it is settled by actual usage, and any excess hold is released automatically.

Model prices may change as upstream providers adjust their pricing. Changes are announced in advance, and completed historical calls are unaffected.

Card and online payments are processed by our payment provider Dodo Payments as the Merchant of Record. Your receipt / invoice is available in the payment confirmation email Dodo sends after checkout. For help, contact [email protected].

For any issue — top-up not credited, an API key that won't work, balance or billing questions, request errors — email [email protected] with your account email, when it happened, and any error message. We'll look into it as soon as we can.

Enterprise

Custom pricing, volume discounts, private deployment, and priority support — tell us about your use case.

  • Concurrency scales with your real usage — uncapped concurrency supported
  • Custom pricing and billing multipliers
  • Priority support and private deployment options
Contact us
Get startedView model prices