Models and pricing
Each divyam-as/* model matches the quality of its equivalent frontier model at 50% of that model's price, thanks to Divyam's routing.
Choose a model
- Open Models & Pricing in the console.
- Pick the model whose frontier equivalent you use today. Each card shows a description, the Context window and the rates.
- Select Use this model to see an example in Getting Started, or copy the full model ID, such as
divyam-as/gpt-6-sol, into your application.
To change models, change the model value your application sends. The same key and base URL work for every model. An ID that isn't in the catalog fails.
Rates
Each card shows rates per million tokens, in your account's currency: Input, Output, Cache read and Cache write. Divyam bills cache reads and cache writes at their own rates. Reasoning tokens count as output tokens and use the Output rate. Prices change, so check them in the console. For how your tokens split, see Cached tokens.
Requested and routed models
Billing uses the rates of the model you request. Divyam may answer with a different backing model, but the response's model field still shows the requested one. Usage → Requested against routed shows the split. See Logs and usage.