Ollama Configuration
This section explains how to configure a local Ollama model in AskSenseOps.
Prerequisites
Before configuring the model, ensure that Ollama is installed and running on the machine hosting the Ollama server. You should also have the model you want to use available in Ollama.
Adding an Ollama Model
- Open AskSenseOps.
- Navigate to the model configuration section.
- Click Add Model.
- Enter the required configuration details.
- Test the configuration and save the model.
Basic Info
| Field | Description | Required |
|---|---|---|
| Display Name | Enter a name to identify the model in AskSenseOps. | Yes |
| Provider | Select Ollama (Local). | Yes |
Credentials
| Field | Description | Required |
|---|---|---|
| API Key | Enter an API key when connecting to Ollama Cloud. Leave this field empty when connecting to a local Ollama server. | No |
Model Configuration
| Field | Description | Required |
|---|---|---|
| Model Name / ID | Enter the name or identifier of the model available in Ollama. | Yes |
| Ollama Server URL | Enter the URL of the Ollama server. The default example is http://localhost:11434. |
Yes |
For a local Ollama server, use the server URL that AskSenseOps can reach. If Ollama is running on another machine, use that machine's reachable address instead of localhost.
Performance
The Performance section allows you to configure how the model generates responses.
| Field | Description | Required |
|---|---|---|
| Compute Mode | Select the compute mode. The default is Auto, which uses the GPU if available. | No |
Context Window (num_ctx) |
Specify the context window size. Leave blank to use the model's maximum. | No |
Max Output Tokens (num_predict) |
Specify the maximum number of output tokens. Leave blank for unlimited output. | No |
| Temperature | Specify the temperature for response generation. Leave blank to use the model's default. | No |
Testing and Saving the Model
After entering the configuration details:
- Review the display name, provider, model identifier, and Ollama Server URL.
- Configure the optional performance settings if required.
- Click Test to verify the configuration.
- If the test succeeds, click Save Model to save the configuration.
If the test fails, verify that the Ollama server is running, that the server URL is reachable from AskSenseOps, and that the specified model is available on the server.