https://github.com/notcoffee418/ollama-terminal
Terminal access to ollama API-only models designed to used with bashrc alias
https://github.com/notcoffee418/ollama-terminal
Last synced: 5 months ago
JSON representation
Terminal access to ollama API-only models designed to used with bashrc alias
- Host: GitHub
- URL: https://github.com/notcoffee418/ollama-terminal
- Owner: NotCoffee418
- License: mit
- Created: 2023-11-04T09:02:22.000Z (over 2 years ago)
- Default Branch: main
- Last Pushed: 2024-12-11T23:33:25.000Z (over 1 year ago)
- Last Synced: 2025-01-23T08:16:17.349Z (over 1 year ago)
- Language: Go
- Size: 21.5 KB
- Stars: 2
- Watchers: 2
- Forks: 0
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# ollama-terminal
## Installation
ollama-terminal is a terminal application that uses [ollama](https://ollama.ai) to generate a prompt through the terminal, allowing you to use non-text models in the terminal.
You can specify an alias for the application, and the model to use for the prompt.
**Features on request**
- [ ] Customizable preprompt
- [ ] Chat finetuning
## Install and update
```bash
curl -s https://raw.githubusercontent.com/NotCoffee418/ollama-terminal/main/install.sh | sudo bash
source ~/.bashrc
```
## Usage
```bash
llama [optional initial prompt]
```
See Configuration section for changing `llama` to a different alias.
## Configuration
The alias is the command you run to activate the application. By default the alias is `llama`.
To change the alias run the following command:
```bash
/usr/local/bin/ollama-terminal set alias [new alias]
source ~/.bashrc
```
To set the model ollama-terminal uses to generate the prompt run the following command:
```bash
/usr/local/bin/ollama-terminal set model [model name]
source ~/.bashrc
```
You can find a list of usable models on the [ollama website](https://ollama.ai/library).