Skip to main content
Vantage integrates with your Fireworks AI account using an API key to ingest cost and usage data through Fireworks AI’s billing summary API. Vantage retrieves daily rated cost and usage data broken down by product (Serverless Inference, Serverless Training API, On Demand Deployments, and Managed Training), the model bucket or accelerator type driving each charge, and the usage category (for example, LLM input tokens or GPU-hours). Usage data is available for Fireworks AI, measured in tokens for inference and training, and in GPU-hours for on-demand deployments.

​
Fireworks AI Permissions

Vantage connects to Fireworks AI using an API key and validates access by listing the Fireworks accounts associated with that key. Vantage cannot deploy models, trigger inference or training jobs, or perform any actions that incur costs in your Fireworks AI account. It’s best practice to create a dedicated API key for Vantage and rotate it according to your organization’s security policy. Vantage does not ingest prompt or response content from Fireworks AI. Only billing metadata, such as token and GPU-hour consumption, is stored.

​
Connect Your Fireworks AI Account

​
Prerequisites

  • You must have a Vantage Organization Owner or Integration Owner role to add or remove this integration. See the Role-Based Access Control documentation for details.
  • You need access to a Fireworks AI account that can create API keys.
  • Create a free Vantage account, then follow the steps below to integrate Fireworks AI costs.

​
Create the Connection

1
Sign in to the Fireworks AI dashboard.
2
On the left navigation, click Settings > API Keys.
3
Click Create API Key > API Key.
4
Give the key a descriptive name, then click Generate Key. You can optionally set an expiration date. If the key expires, Vantage can no longer refresh your Fireworks AI cost data, so you will need to generate a new key and update it in Vantage before then.
5
Copy the API key.
Fireworks AI only shows the API key once. Save it securely before closing the dialog.
6
From the top navigation in Vantage, click Settings.
7
On the left navigation, select Integrations > Fireworks AI.
8
The Fireworks AI integrations page is displayed. Ensure you are on the Connect tab.
9
Click Add API Key and complete the integration form:
  • API Key: Paste the Fireworks AI API key.
  • Description: Enter a label to identify this Fireworks AI connection in Vantage. This value is used as the Organization dimension in Cost Report filters. Fireworks accounts accessible with this key are connected automatically.
10
Click Connect Account.
After clicking Connect Account, you will see the status of your integration change to Importing within the Vantage console. This status indicates that Vantage is actively importing your Fireworks AI cost data. Vantage backfills up to six months of historical cost data upon initial connection. See the Integration Status documentation for details on integration statuses. You can connect multiple Fireworks AI accounts, each with its own API key, as separate integrations in Vantage.
As soon as costs are processed, they will be available on your All Resources Cost Report. If you decide to remove your Fireworks AI integration from Vantage, all costs associated with that Fireworks AI account will be removed from the Vantage console.
Vantage shows gross rated usage cost from Fireworks AI’s billing summary. Credits, discounts, taxes, and invoice adjustments are not currently exposed by Fireworks AI’s billing summary API, so costs in Vantage may differ from your final Fireworks AI invoice. If you have negotiated pricing you’d like reflected, contact support@vantage.sh.

​
Next Steps - Manage Workspace Access

Once the import is complete and the integration status changes to Stable, you can select which workspaces this integration is associated with. See the Workspaces documentation for information.

​
Data Refresh

See the provider data refresh documentation for information on when data for each provider refreshes in Vantage. Fireworks AI data refreshes once daily.

​
Fireworks AI Reporting Dimensions

On Fireworks AI Cost Reports, you can filter and group across several dimensions:
  • Organization (the description you enter when connecting the integration)
  • Service (i.e., the Fireworks AI product, such as Serverless Inference, Serverless Training API, On Demand Deployments, or Managed Training)
  • Category (i.e., the model bucket, accelerator type, or base model name driving the charge, such as GLM 5.2 or NVIDIA_B200_180GB)
  • Subcategory (i.e., the usage category, such as LLM input tokens (uncached), LLM output tokens, or On-demand deployment GPU-hours)
  • Charge Type (Usage)
In addition, Fireworks AI costs include the following provider tag, which is automatically created by Vantage and available for filtering and grouping alongside any Virtual Tags you create: