Ollama â
Ollama is an open-source client for running open-source models offline locally. It supports all computer platforms and most open-source models.
This guide introduces how to configure and use the Ollama model in Aide.
You can find more detailed information in the Ollama Official Documentation.
Install Ollama â
Before using Ollama, please ensure you have downloaded and installed the client from the Ollama Official Website. You can download it from this link. For more information, please refer to the Ollama Open-Source Repository.
API Base URL Configuration â
You need to configure aide.openaiBaseUrl to http://localhost:11434/v1
API Key Configuration â
You can configure aide.openaiKey with any value. For example, you can use sk-ollama.
Model Configuration â
You need to configure aide.openaiModel to the Ollama model. We recommend using the llama3.1 model. For more models, please refer to the Supported Models List.
Example Configuration File â
Below is a complete configuration example:
{
"aide.openaiBaseUrl": "http://localhost:11434/v1",
"aide.openaiKey": "sk-ollama",
"aide.openaiModel": "llama3.1"
}