Your own models and keys

Besides the platform catalog, a workspace can bring its own models: an OpenAI-compatible provider connection stores a base URL, a key and the list of models you allow. One key serves models of all three kinds — chat, embeddings and rerank.

Connecting a provider

  1. Open Settings → Integrations and create an OpenAI-compatible provider connection.
  2. Fill in the Base URL (for example https://api.openai.com/v1) and the API key.
  3. Press Load from provider — the platform asks that endpoint for its model list. If the provider does not serve one, use Add manually.
  4. Turn on the models you want: loaded models arrive disabled.
  5. Pick the Model purpose on each row — Chat, Embeddings or Rerank.
  6. Save the connection.
app.iterna.ai

Integrations

OpenAI-compatible provider

Check connectionCreate
Base URL
https://api.openai.com/v1
API key*
••••••••••••
Models
on gpt-4o Chat Images · Audio on text-embedding-3-small Embeddings off rerank-v3.5 Rerank

Text is always accepted; the modality chips restrict which attachments are sent to the model.

Models inside one connection: purpose is chosen per row.

How a model reaches the dropdowns

A connection model gets an id of the form ws:<connection id>:<model id> and sits next to the catalog models — but only in the dropdown that matches its purpose:

PurposeWhere it appearsExample label
Chatthe Model field of the llm nodeAcme OpenAI · gpt-4o
Embeddingsthe embedding node, knowledge-base creationAcme OpenAI · text-embedding-3-small
Rerankthe rerank nodeAcme OpenAI · rerank-v3.5

The label is “connection name · model label”, so two connections carrying the same model stay distinguishable. In Settings → Models these rows live under the workspace group, and their visibility is toggled there too.

Fallback chains

A separate connection kind is the fallback chain. It is assembled from models you already connected, listed in the order they are tried, and appears in the LLM dropdown as a single entry. Checkboxes decide what makes the chain advance — request timeouts, HTTP 4xx, HTTP 5xx.

Media nodes and the web-search node take a different route: they run through OpenRouter, so your own key is supplied by an OpenRouter connection and picked in the node’s connection field. Leave that field empty and the node uses the platform key.

What this does to billing

Your model runs on your key and the provider bills you, so the platform does not meter its tokens — otherwise the same work would be paid for twice.

ChargeCatalog modelYour model (BYO)
Workflow run1 credit1 credit
Each LLM node1 credit1 credit
LLM tokensthe model's tariffnot billed
Embedding and rerank tokensbilledno ledger row at all

The tokens of a BYO call stay visible under Usage — the node really did run — but their price is zero.

What next