Pricing · Spring promo
List price visible. Promo price honest.
Every model shows both the provider list price and our promo rate. Recharge prepaid USD in the customer console, create an API key, switch models by name.
Model rates · per 1M tokens
Three live routes. Two on promo.
Promo prices apply through 5/31; rates revert to list afterwards. DeepSeek stays at its always-on low rate.
| Model | Input | Output | Status | Use case |
|---|---|---|---|---|
| gpt-5.5 OpenAI · GPT-5.5 (1M context) |
$5.00$3.50-30% | $30.00$21.00-30% | Live · Promo | Multimodal, general reasoning, coding, tool use |
| claude-opus-4-7 Anthropic · Claude Opus 4.7 |
$15.00$11.25-25% | $75.00$56.25-25% | Live · Promo | Long-context reasoning, careful coding, agents |
| deepseek-chat DeepSeek Chat |
$0.30 | $0.60 | Live | Always-on low rate, chat, coding help, OpenAI SDK migration |
All prices in USD per 1M tokens. Promo pricing through 5/31; afterwards rates revert to list. ModelBridge does not apply credit multipliers — the rates above are what your prepaid balance is debited.
Credits
Recharge in the console, not a public test page.
The public site explains packages. The customer console owns checkout, balance, API keys, usage, and invoices.
Operator
Custom
Higher volume, route planning, and upcoming Chinese multimodal coverage.
Review docs
Formal customer flow: open the console, register or sign in, recharge credits, create an API key, then call
https://api.modelbridgeapi.com/v1. The old standalone dashboard checkout is no longer part of the customer path.