# everruns providers

> LLM providers and their credentials. CLI reference for everruns providers.

Source: <https://everruns.com/docs/reference/cli/providers/>

LLM providers and their credentials.

| Command                                                                   | What it does                                                              |
| ------------------------------------------------------------------------- | ------------------------------------------------------------------------- |
| [`providers create`](#providers-create)                                   | Create a new LLM provider.                                                |
| [`providers delete`](#providers-delete)                                   | Delete an LLM provider.                                                   |
| [`providers get`](#providers-get)                                         | Get a specific LLM provider.                                              |
| [`providers list`](#providers-list)                                       | List all LLM providers.                                                   |
| [`providers update`](#providers-update)                                   | Update an LLM provider.                                                   |
| [`providers check-credentials check`](#providers-check-credentials-check) | Check a provider API key without storing it.                              |
| [`providers models create`](#providers-models-create)                     | Create a new model for a provider.                                        |
| [`providers models list`](#providers-models-list)                         | List models for a specific provider.                                      |
| [`providers models review`](#providers-models-review)                     | Mark a provider’s discovered models as reviewed, clearing their new flag. |
| [`providers sync-models sync`](#providers-sync-models-sync)               | Discover and sync models from a provider.                                 |

## providers create

Create a new LLM provider.

```bash
everruns providers create [OPTIONS] --name <name> --provider-type <provider_type>
```

| Flag                                  | Description                    |
| ------------------------------------- | ------------------------------ |
| `--api-key <API_KEY>`                 | Credential for the provider.   |
| `--base-url <BASE_URL>`               | Custom API endpoint.           |
| `--name <NAME>`                       | Required. Human-readable name. |
| `--provider-type <PROVIDER_TYPE>`     | Required. LLM provider type.   |
| `--request-options <REQUEST_OPTIONS>` |                                |
| `--trace <TRACE>`                     |                                |

Example:

```bash
# Connect an LLM provider with its API key
everruns providers create --name openai-prod --provider-type openai --api-key "$OPENAI_API_KEY" --reason 'Enable OpenAI models'
```

## providers delete

Delete an LLM provider.

```bash
everruns providers delete [OPTIONS] [ID]
```

| Flag        | Description                 |
| ----------- | --------------------------- |
| `--id <ID>` | Prefixed public identifier. |

Example:

```bash
# Remove a provider and stop using its models
everruns providers delete provider_01h9 --reason 'Account closed'
```

## providers get

Get a specific LLM provider.

```bash
everruns providers get [OPTIONS] [ID]
```

| Flag        | Description                 |
| ----------- | --------------------------- |
| `--id <ID>` | Prefixed public identifier. |

Example:

```bash
# Show a provider's type and status
everruns providers get provider_01h9
```

## providers list

List all LLM providers.

```bash
everruns providers list [OPTIONS]
```

Example:

```bash
# See which LLM providers are connected
everruns providers list
```

## providers update

Update an LLM provider.

```bash
everruns providers update [OPTIONS] --id <id>
```

| Flag                                  | Description                           |
| ------------------------------------- | ------------------------------------- |
| `--api-key <API_KEY>`                 | Replacement credential.               |
| `--base-url <BASE_URL>`               | Replacement API endpoint.             |
| `--id <ID>`                           | Required. Prefixed public identifier. |
| `--name <NAME>`                       | Human-readable name.                  |
| `--provider-type <PROVIDER_TYPE>`     |                                       |
| `--request-options <REQUEST_OPTIONS>` |                                       |
| `--status <STATUS>`                   |                                       |
| `--trace <TRACE>`                     |                                       |

Example:

```bash
# Rotate a provider's API key
everruns providers update --id provider_01h9 --api-key "$OPENAI_API_KEY" --reason 'Scheduled key rotation'
```

## providers check-credentials check

Check a provider API key without storing it.

```bash
everruns providers check-credentials check [OPTIONS] --api-key <api_key> --provider-type <provider_type>
```

| Flag                              | Description                        |
| --------------------------------- | ---------------------------------- |
| `--api-key <API_KEY>`             | Required. The credential to probe. |
| `--base-url <BASE_URL>`           | Optional custom endpoint.          |
| `--provider-type <PROVIDER_TYPE>` | Required. LLM provider type.       |

Example:

```bash
# Verify an API key works before saving a provider
everruns providers check-credentials check --provider-type openai --api-key "$OPENAI_API_KEY"
```

## providers models create

Create a new model for a provider.

```bash
everruns providers models create [OPTIONS] --display-name <display_name> --model-id <model_id> --provider-id <provider_id>
```

| Flag                            | Description                                                                              |
| ------------------------------- | ---------------------------------------------------------------------------------------- |
| `--capabilities <CAPABILITIES>` | Capability labels for the model. Comma-separated or a JSON array of strings. Repeatable. |
| `--display-name <DISPLAY_NAME>` | Required. Human-readable display name.                                                   |
| `--enabled`                     | Whether this resource is enabled.                                                        |
| `--is-favorite`                 | Show the model first in pickers.                                                         |
| `--model-id <MODEL_ID>`         | Required. LLM model’s prefixed public identifier.                                        |
| `--profile-key <PROFILE_KEY>`   | Stable model profile to assign to the model.                                             |
| `--provider-id <PROVIDER_ID>`   | Required. LLM provider’s prefixed public identifier.                                     |
| `--service <SERVICE>`           |                                                                                          |

Example:

```bash
# Register a model that provider discovery does not list
everruns providers models create --provider-id provider_01h9 --model-id gpt-5.1 --display-name 'GPT-5.1' --enabled true --reason 'Make the new model selectable'
```

## providers models list

List models for a specific provider.

```bash
everruns providers models list [OPTIONS] --provider-id <provider_id>
```

| Flag                          | Description                                          |
| ----------------------------- | ---------------------------------------------------- |
| `--provider-id <PROVIDER_ID>` | Required. LLM provider’s prefixed public identifier. |

Example:

```bash
# See which models one provider offers
everruns providers models list --provider-id provider_01h9
```

## providers models review

Mark a provider’s discovered models as reviewed, clearing their new flag.

```bash
everruns providers models review [OPTIONS] --id <id>
```

| Flag        | Description                           |
| ----------- | ------------------------------------- |
| `--id <ID>` | Required. Prefixed public identifier. |

Example:

```bash
# Clear the new flag once you have looked at a provider's newly discovered models
everruns providers models review --id provider_01h9 --reason 'Checked new models'
```

## providers sync-models sync

Discover and sync models from a provider.

```bash
everruns providers sync-models sync [OPTIONS] --id <id>
```

| Flag        | Description                           |
| ----------- | ------------------------------------- |
| `--id <ID>` | Required. Prefixed public identifier. |

Example:

```bash
# Pull a provider's current model list after it ships new models
everruns providers sync-models sync --id provider_01h9 --reason 'New models released'
```