Settings
Model providers¶
The key or endpoint a run's model calls go to.
| Provider | Setting | Endpoint |
|---|---|---|
| TypeSafe AI | TYPESAFE_API_KEY |
System One judgement only |
| OpenAI | OPENAI_API_KEY |
api.openai.com |
| OpenRouter | OPENROUTER_API_KEY |
https://openrouter.ai/api/v1, fixed |
| Anthropic | ANTHROPIC_API_KEY |
Standard runs only, through the relay |
| Local or self-hosted | MODEL_API_URL, optional MODEL_API_KEY |
An OpenAI-compatible endpoint, API path included |
Configure a provider¶
- Settings → Model providers → + Add: choose a company from the searchable catalogue, then enter its key or endpoint. Fields save automatically.
- Configured providers appear in a compact list. Edit opens their connection fields; Done returns to the list.
- In the builder's Model step, choose the provider, then search its model list or enter an exact model ID. See model selection.
A local endpoint¶
Use the Settings page to save these values, or merge the shown properties into
your existing settings.json (location and precedence).
settings.json · local endpoint
{
"MODEL_API_URL": "http://127.0.0.1:8000/v1",
"MODEL_API_KEY": ""
}
| Rule | |
|---|---|
| No key | Leave it empty; Nullius sends the placeholder local |
| Reachable from | The calling machine; loopback on a run host is that host's |
Where credentials go¶
| A run | Its provider's key and endpoint, as environment, never written |
| A recipe, a log | Nothing; logs are redacted |
| A standard run on Anthropic | The relay alone |
-
Precedence, reloads.
-
The judges.
-
An agent's vendor.