Today, Vantage is launching support for Together AI costs, giving customers the ability to track inference and GPU spend alongside the rest of their cloud and AI providers. Once connected, Vantage automatically imports daily Together AI costs by product, model, GPU, project, and endpoint, so teams can monitor dedicated GPU, serverless inference, and fine-tuning spend, set budgets, detect anomalies, and allocate Together AI costs in Cost Reports and Virtual Tags.

Together AI is an inference platform teams use for open-model serving, dedicated GPU endpoints, and fine-tuning, often next to cloud providers like AWS, GCP, or Azure. Previously, that spend lived in Together cost analytics, separate from the rest of an AI budget. That made it harder to see which models, GPUs, projects, and endpoints were driving cost, and to include Together AI in the same reports and allocation rules used for everything else.

Now, customers connect a Together AI organization with an API key, and Vantage brings in the rated costs Together AI reports for that organization. Dedicated Inference is broken out by GPU type and on-demand or reserved GPU-hours. Serverless Inference is broken out by model and by input, cached input, and output tokens. Costs are enriched with Managed AI Tags for the model, token type, API key, and endpoint, so teams can compare Together AI with other AI providers in the same report. Customers can include Together AI costs in Cost Reports, Budgets, Anomaly Detection, and more.

The Together AI integration is now available to all Vantage customers. To connect, go to the Integrations page in account settings and select Together AI. You can learn more via the Together AI section in Vantage's documentation.

Frequently Asked Questions

1. What is being launched today?

Vantage is launching native support for Together AI costs. Customers can connect a Together AI organization with an API key so Vantage automatically imports daily costs by product, GPU, model, project, and endpoint, and surfaces them alongside the rest of their cloud and AI costs.

2. Who has access to this integration?

The Together AI integration is available to all Vantage customers across all subscription tiers.

3. How much does this integration cost?

There is no additional fee for using the integration. However, Together AI costs will be included in quota tier enforcement. If your Together AI costs push you over your current tier limit, you may be prompted to upgrade. See the Pricing page for details.

4. How do I set up the integration?

Create an API key in Together AI, then add it in Vantage:

1. In Together AI, open the project where you want to create the key and go to that project's API keys settings.
2. Select Create API Key, give it a name, and copy it. Together shows the key only once.
3. In Vantage, go to the Integrations page in account settings and select Together AI.
4. Paste the key into the API Key field, enter a Description to identify the connection, and click Connect Account.

Vantage checks that the key can read the organization's Billing Usage API before it begins importing costs. The API returns costs for the key's entire organization, so connect one key per organization. To connect another organization, click Add Another API Key and repeat these steps.

5. What permissions are required on the Together AI side?

You need a Together AI project API key for an organization that can read the Billing Usage API. Project admins and members can create keys. Together does not offer a billing-only key. See Together's API key documentation for how keys are scoped.

6. What permissions are needed within Vantage?

You must have a Vantage Owner or Integration Owner role to add or remove the Together AI integration.

7. What permissions does Vantage have in my Together AI account?

A Together AI project key can also run inference and spend credits. Vantage only reads the Billing Usage API, and will never create endpoints, run inference or fine-tuning, or perform any other actions.

8. What dimensions can Together AI costs be filtered and grouped by?

Together AI costs support aggregating and filtering on:

• Organization (the Together organization ID)
• Account (the Together project ID, when Together includes one)
• Service (the product, such as Dedicated Inference, Serverless Inference, GPU Clusters, Fine-Tuning, Code Sandbox, Managed Storage, or Provisioned Throughput)
• Category (the GPU type, such as H100-80GB, the model name, or the product name)
• Subcategory (the usage class, such as On-demand GPU-hours, Reserved GPU-hours, Input tokens (cached), Input tokens (uncached), or Output tokens)
• Resource (the endpoint name, when the line item has one)
• Charge Type (Usage)
• Tags, including together:capacity_type (BASE or BURST) and Managed AI Tags for model, token type, API key, and endpoint

9. Can I view usage data for Together AI?

Yes. Dedicated Inference usage is reported in GPU-hours. Serverless Inference usage is reported in tokens, split into cached input, uncached input, and output. Products such as fine-tuning, storage, clusters, and code sandbox do not include a usage unit.

10. How are credits, discounts, and tax handled?

Vantage imports the rated cost Together reports for each line item. Credits, discounts, and tax are not subtracted, so costs in Vantage can differ from the final Together AI invoice. Days with no line items are not imported.

11. Are there Active Resources for Together AI?

No. Active Resources are not available for Together AI.

12. How often does Together AI data refresh in the Vantage console?

Together AI data refreshes daily in the Vantage console. During the first five days of each month, Vantage also refreshes the previous month so late-arriving charges are captured. Daily costs run through yesterday, UTC.

13. How far back will the data go?

On initial connection, Vantage imports up to six months of historical Together AI costs.

14. Can I connect multiple Together AI organizations?

Yes. Each API key returns billing for its entire organization, including every project, so connect one key per organization. A second key for the same organization imports those costs again.

15. What happens if I remove the Together AI integration?

If you remove your Together AI integration from Vantage, all associated costs will be removed from the Vantage console. Removing the integration does not revoke the API key in Together AI.

16. Does Vantage store prompts or model outputs?

No. Vantage only imports billing data, such as costs by product, model, GPU, project, and endpoint. It does not access prompts or model outputs.