Linking Together AI as a source
This source is currently in alpha. The interface and available tables may change.
The Together AI connector syncs your fine-tuning jobs, batch inference jobs, uploaded files, dedicated endpoints, evaluations, and the model catalog into the PostHog Data warehouse, so you can analyze your AI training and inference operations alongside your product data.
Prerequisites
You need a Together AI account with an API key. Together AI requires a payment method on file before API keys can make requests.
Adding a data source
- In PostHog, go to the Sources tab of the data pipeline section.
- Click + New source and click Link next to this source.
- Enter your credentials (see Configuration below) and click Next.
- Select the tables you want to sync, choose a sync method and frequency, then click Import.
Once the syncs are complete, you can start querying this data in PostHog.
When linking Together AI, you'll need:
- API key – find or create one under Settings → API keys in your Together AI account.
Sync modes
Each table can be synced in one of several modes, depending on what the source supports:
- Webhook (when available) – the source pushes changes to PostHog in real time. Fastest freshness, lowest ongoing cost, and the only mode that reliably captures updates and deletes.
- Incremental – only new or updated rows are synced on each run, using a cursor field (such as an
updated_attimestamp). Cheaper than a full refresh, but deletes aren't captured. - Append only – new rows are appended using a cursor field; existing rows are never updated. Ideal for immutable, append-only tables like event logs.
- Full refresh – the whole table is reloaded on every sync. Use it when a table has no reliable cursor or when you need deletions reflected.
See sync methods for a full explanation of how each mode works and how to choose between them.
Together AI's list endpoints don't support timestamp filtering, so all tables sync with full refresh.
Configuration
| Option | Type | Required |
|---|---|---|
API key | password | Yes |
Supported tables
| Table | Description | Sync method | Incremental field | Primary key |
|---|---|---|---|---|
fine_tunes | A fine-tuning job: the base model, training data, hyperparameters, progress counters, and price of each fine-tune run on Together AI. | Full refresh | — | id |
batches | A batch inference job: asynchronous bulk processing of requests from an uploaded input file against a model, with progress and output file references. | Full refresh | — | id |
files | Files uploaded to Together AI, used as training data for fine-tuning or as inputs/outputs of batch inference jobs. | Full refresh | — | id |
endpoints | Dedicated model endpoints deployed by the account: the model, hardware configuration, autoscaling settings, and current state of each deployment. | Full refresh | — | id |
evaluations | Model evaluation jobs: classify, score, or compare runs that judge model outputs, with their parameters and aggregate results. | Full refresh | — | workflow_id |
models | The Together AI model catalog: every serverless model available to the account, with type, context length, and per-token pricing. | Full refresh | — | id |
Troubleshooting
- If you see an authentication error, your API key is invalid or has been revoked. Create a new key in your Together AI settings, then reconnect.
- Usage and billing data isn't available through Together AI's public API, so it can't be synced. It's only visible in the Together AI dashboard.
If your sync is failing or data looks wrong, see the Data warehouse troubleshooting guide. If that doesn't help, contact support – we're happy to help.