Connect Fireworks AI
Fireworks AI is a generative AI inference platform for running, fine-tuning, and deploying open models. The Fireworks AI integration imports your daily rated cost and usage data into DoiT Cloud Intelligence so you can analyze Fireworks AI spend in Cloud Analytics.
Prerequisites
-
A Fireworks AI account with administrator access.
-
A DoiT account with the Billing Profiles Admin permission.
Required credentials
| Field | Description | Where to find it |
|---|---|---|
| Account ID | Your Fireworks AI account identifier | Fireworks AI dashboard > Account Settings |
| API Key | An admin API key | Fireworks AI dashboard > Settings > API Keys > Create API Key |
The API key must belong to an account administrator. Cost attribution by user and by API key uses a Fireworks AI endpoint that requires account administrator access. With a non-admin key, DoiT still imports your overall billing data, but per-user and per-key attribution data is skipped.
Create an API key
-
Sign in to the Fireworks AI dashboard as an account administrator.
-
Navigate to Settings > API Keys.
-
Select Create API Key and give it a descriptive name.
-
Copy the key.
CautionThe key is only shown once. Make sure to copy it before closing the dialog.
Connect your Fireworks AI account
-
Sign in to Cloud Intelligence, select Data ingestion and integrations > Integrations from the top navigation mega menu.
-
Search for Fireworks AI in the integrations catalog.
-
Select Connect + on the integration card.
-
Enter your Account ID and API Key.
-
Select Test connection to verify DoiT can access your Fireworks AI account.
-
If the test is successful, select Save to complete the setup.
If the test fails or you need further assistance, please open an expert inquiry.
Upon initial connection, DoiT imports up to 180 days of historical Fireworks AI cost data. The import process can take a few hours. You'll receive an email when the data is available in Cloud Analytics.
Data refresh
DoiT refreshes your Fireworks AI billing data every six hours. Fireworks AI reports costs at daily granularity; a day's data is typically final about one day after it is generated.
Reporting dimensions
Once connected, Fireworks AI data maps to the following standard Cloud Analytics dimensions:
| Cloud Analytics dimension | Fireworks AI value | Example |
|---|---|---|
| Provider | Fireworks AI | Fireworks AI |
| Billing Account | Account ID | my-account |
| Service | Product line | Serverless Inference, Dedicated Deployments, Fine-tuning |
| SKU | Model, or accelerator type for dedicated deployments | GPT OSS 120B, NVIDIA_B300_288GB |
In addition to standard dimensions, the following Fireworks AI-specific labels are available:
| Label | Description |
|---|---|
| Model | The model that generated the charge |
| Deployment Type | Product line: SERVERLESS, DEDICATED_DEPLOYMENT, or TRAINING |
| Usage Type | Usage category (e.g., LLM input tokens (uncached), Batch Inference Output Tokens) |
| Cached | Whether the input tokens were served from cache (true or false) |
| Accelerator | GPU type for dedicated-deployment charges (e.g., NVIDIA_B300_288GB) |
| Account | The Fireworks AI account the charge belongs to |
Usage amounts are measured in tokens for serverless inference and fine-tuning, and in GPU-hours for dedicated deployments.
Fireworks AI costs appear in the GenAI Intelligence dashboard alongside other supported GenAI providers.
Imported costs reflect gross rated usage. Credits, discounts, taxes, and invoice adjustments are not exposed by Fireworks AI's billing API and are therefore not included.
Delete your Fireworks AI connection
-
Sign in to Cloud Intelligence, select Data ingestion and integrations > Integrations from the top navigation mega menu.
-
Search for Fireworks AI in the integrations catalog.
-
Select Manage on the integration card.
-
Select the kebab menu (⋮) at the rightmost end of the account you want to remove and select Delete.
-
Select Delete to confirm.