> For the complete documentation index, see [llms.txt](https://docs.apismart.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.apismart.ai/models-and-pricing/model-pricing-and-billing.md). # Model Pricing and Billing ApiSmart deducts API usage costs from your wallet balance. The final cost depends on the selected model, billing unit, request configuration, and actual usage. Before integrating a model, review its pricing on the [**Models**](https://www.apismart.ai/models) page and in the **Price Configuration** section of the model details page. *** ### 💰 Review the Billing Mode The Models page may display billing labels such as: * Pay-per-use * Billing per use * Tiered pricing * Time-based billing Billing Mode provides a general indication of how a model is charged. Always review the model’s **Price Configuration** for the exact billing unit and rate. > ⚠️ **Important:** Do not rely on the Billing Mode label alone when estimating request costs. *** ### Understand Common Billing Units #### Per Million Units Some prices are displayed with `/M`: ``` Text generation: $10.4/M Video input: $6.3/M ``` `/M` indicates a price per one million billable units. However, the type of unit may depend on the model and input type. Do not assume that every `/M` value represents standard text Tokens. Review the model’s **Price Configuration** and details when the billing unit is unclear. *** #### Per Image Some image generation models are billed for each generated image: ``` Billing: $0.04/image ``` For example, generating five images at this rate would cost: ``` 5 × $0.04 = $0.20 ``` The final cost may also depend on model-specific settings. *** #### Per Generated Second Some video models use time-based billing: ``` 720P: $0.145/s 1080P: $0.24/s ``` `/s` means the price applies per generated second. For an eight-second video at `$0.145/s`: ``` 8 × $0.145 = $1.16 ``` Higher resolutions may use higher rates. *** #### Cache Pricing Some language models may display separate **Cache hit** and **Cache creation** rates. These rates apply only when the selected model supports the corresponding caching behaviour and may differ from standard input or output pricing.
*** ### Review Configuration-Based Pricing Some models display different prices based on configuration, such as: * Resolution * Input type * Generated duration * Reference media * Model-specific options For example: ``` 720P: $0.145/s1080P: $0.24/s ``` The `default` option represents the model’s default configuration and does not necessarily mean the lowest-cost option. *** ### Where to Check Pricing #### Model Card Use the model card for a quick pricing comparison between models. It may display: * Billing Mode * Pricing summary * Model category * Capability tags
#### Model Details Select **View Details** to review the model’s complete **Price Configuration** and supported settings.
Use the Model Details page as the primary pricing reference before running a production workload. *** ### Estimate Request Costs | Billing Type | Simplified Estimate | | ----------------- | ------------------------------------------------ | | Per million units | `Billable Units ÷ 1,000,000 × Price per Million` | | Per image | `Number of Images × Price per Image` | | Per second | `Generated Duration × Price per Second` | Example: ``` Billable Units: 100,000Price: $10/M100,000 ÷ 1,000,000 × $10 = $1 ``` These are simplified estimates. Actual costs may vary based on the selected model, configuration, input type, retries, or model-specific billing rules. *** ### Wallet Balance and API Token Limits Wallet balance and API Token usage limits are separate controls. For example: ``` Wallet Balance: $49API Token Limit: $5 ``` Even with $49 available in the wallet, this Token can consume only its configured $5 allowance. If a request fails because of available funds or allowance, check both: * **Wallet → Current Balance** * **API Tokens → Used / Remaining** Review both **Wallet** and **API Tokens** before adding more funds. *** ### 📊 Review the Final Request Cost After making a request: 1. Open **Console → Usage Logs**.
2. Locate the request. 3. Confirm the **Model**, **Consumption**, and **Cost**. 4. Compare the recorded cost with your estimate if needed. For complete log fields, filtering, and troubleshooting, see **Usage Logs and Costs**. *** ### Control Production Costs Before running a high-volume workload: * Review the current model pricing. * Confirm the billing unit and configuration. * Run several representative test requests. * Review the actual cost in **Usage Logs**. * Estimate expected monthly usage. * Set an appropriate API Token usage limit. For example: ``` Average Request Cost: $0.02 Monthly Requests: 10,000 Estimated Monthly Cost: $0.02 × 10,000 = $200 ``` *** ### 🛠️ Troubleshoot Unexpected Charges If a request costs more than expected, check: * Model ID * Billing unit * Resolution or generation duration * Number of generated results * Input and output usage * Automatic retries * Background applications or jobs * Other API Tokens using the same wallet Then review the related request in **Usage Logs**. If you see unfamiliar usage, disable the affected **API Token** immediately and review **Usage Logs**. See [**Secure Your API Tokens**](/api-tokens/secure-your-api-tokens.md) for detailed security steps. > 🔐 Never provide your full **API Token** when reporting a billing issue. *** ### 🚀 Next Steps Continue with: * [**Choose a Model and Find Its Model ID**](/models-and-pricing/models-and-model-ids.md) * [**View Your Balance and Transaction Records**](/account-and-wallet/view-your-balance-and-transaction-records.md) * [**View API Usage Logs and Costs**](/api-usage/usage-logs-and-costs.md) * [**Add Funds and Understand Wallet Credits**](/account-and-wallet/add-funds-and-understand-wallet-credits.md) * [**Create and Manage API Tokens**](/api-tokens/create-and-manage-api-tokens.md)