Provider list price, minus your plan discount
You pay per 1M tokens, input and output priced separately, deducted from prepaid credits. No markup on top-ups and no hidden per-request fees.
Starter
Pay as you go from prepaid credits.
$0/month
- 5% below provider list price
- All models in the catalog
- 60 requests per minute per key
- Up to 3 API keys
- Email support, 1 business day
Pro
Most teamsFor teams with regular production traffic.
$49/month
- 15% below provider list price
- 600 requests per minute per key
- Unlimited keys with caps and allowlists
- 5 team members, roles and audit log
- Usage CSV export and API
Enterprise
For high volume and procurement needs.
Custom
- Volume discount set per contract
- Custom rate limits
- SAML 2.0 SSO (Okta, Entra ID)
- Provider routing rules
- Invoice billing, 4-hour support response
Model rates
USD per 1M tokens. List prices are each provider's standard published rate, checked 2026-10-23. When a provider changes its price, our rate follows.
| Model ID | List in / out | Starter in / out | Pro in / out |
|---|---|---|---|
| openai/gpt-6-astra | $10.00 / $50.00 | $9.50 / $47.50 | $8.50 / $42.50 |
| openai/gpt-6.1-sol | $2.00 / $10.00 | $1.90 / $9.50 | $1.70 / $8.50 |
| openai/gpt-6-luna | $0.10 / $0.50 | $0.095 / $0.475 | $0.085 / $0.425 |
| anthropic/claude-fable-5.1 | $10.00 / $50.00 | $9.50 / $47.50 | $8.50 / $42.50 |
| anthropic/claude-opus-5.5 | $4.00 / $20.00 | $3.80 / $19.00 | $3.40 / $17.00 |
| anthropic/claude-sonnet-5.5 | $2.00 / $10.00 | $1.90 / $9.50 | $1.70 / $8.50 |
| anthropic/claude-haiku-5.5Prompts up to 100K tokens | $0.10 / $0.50 | $0.095 / $0.475 | $0.085 / $0.425 |
| google/gemini-4-argonProvider promo price through 2026-11-30 | $2.00 / $10.00 | $1.90 / $9.50 | $1.70 / $8.50 |
| google/gemini-3.8-flashProvider promo price through 2026-12-31 | $0.75 / $3.75 | $0.712 / $3.56 | $0.637 / $3.19 |
| deepseek/deepseek-v4-proPeak-hour, cache-miss rate | $1.32 / $3.96 | $1.25 / $3.76 | $1.12 / $3.37 |
| deepseek/deepseek-flashPeak-hour, cache-miss rate | $0.30 / $1.20 | $0.285 / $1.14 | $0.255 / $1.02 |
- Prompt-cache hits and Batch jobs are billed at the provider's cached or batch rate, minus the same plan discount.
- DeepSeek off-peak rates are half the peak rate shown; peak hours are 01:00–04:00 and 06:00–10:00 UTC, Monday to Friday.
- Fractions of a cent are rounded per request to six decimal places.
How credits work
- Minimum top-up $10
- Pay by card. Pro and Enterprise plan fees are billed monthly, separate from credits.
- No top-up fee
- $100 paid is $100 of credit.
- Credits don't expire
- Unused balance can be refunded as described in our Refund Policy.
- Bring your own key
- Requests sent with your own provider key are billed by that provider and don't use credits.
- Low-balance alerts
- Set a threshold and get an email before your balance runs out.