Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/Renset/macai
Swift powered native macOS client for Ollama, ChatGPT and compatible API-backends
https://github.com/Renset/macai
ai api-client bot chat chatgpt chatgpt-api llm macos ollama swift swiftui
Last synced: about 1 month ago
JSON representation
Swift powered native macOS client for Ollama, ChatGPT and compatible API-backends
- Host: GitHub
- URL: https://github.com/Renset/macai
- Owner: Renset
- License: mit
- Created: 2023-03-21T20:32:29.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2024-05-21T22:52:51.000Z (about 1 month ago)
- Last Synced: 2024-05-21T23:32:48.827Z (about 1 month ago)
- Topics: ai, api-client, bot, chat, chatgpt, chatgpt-api, llm, macos, ollama, swift, swiftui
- Language: Swift
- Homepage: https://renset.gumroad.com/l/macai
- Size: 12.6 MB
- Stars: 68
- Watchers: 4
- Forks: 8
- Open Issues: 2
-
Metadata Files:
- Readme: README.md
- License: LICENSE.md
Lists
- project-awesome - Renset/macai - Swift powered native macOS client for Ollama, ChatGPT and compatible API-backends (Swift)
README
# macai
macai (macOS AI) is a simple yet powerful native macOS client made to interact with modern AI tools (ChatGPT- and Ollama-compatible API are supported).
## Downloads
You can download latest signed binary on [Releases](https://github.com/Renset/macai/releases) page.
You can also support project on [Gumroad](https://renset.gumroad.com/l/macai).
## Build from source
Checkout main branch and open project in Xcode 14.3 or later
## Features
- ChatGPT/Ollama and other compatible API support
- Organized with chats, where each chat has its own context
- Customized system messages (instructions) per chat
- System-defined light/dark theme
- Backup and restore your chats
- Customized context size
- Any model with compatible API can be used
- Formatted code blocks
- Formatted tables, copy as JSON
- With tabs, one can easily work with multiple chats simultaneously
- Data is stored using CoreData
- Streamed responses
- Automatically generate chat names
## Project status
Project is in the active development phase.
## Contributions
Contributions are welcomed. Take a look at [macai project page](https://github.com/users/Renset/projects/1) and [Issues page](https://github.com/Renset/macai/issues) to see planned features/bug fixes, or create a new one.
## Run with ChatGPT
To run macai with ChatGPT, you need to have ChatGPT API token. You can get it [here](https://help.openai.com/en/articles/4936850-where-do-i-find-my-secret-api-key).
Add the token to the settings and you are ready to go.
Note: by default, gpt-3.5-turbo model is selected. You can change it in the New Chat settings.
## Run with [Ollama](https://ollama.com)
Ollama is the open-source back-end for various LLM models.
Run with Ollama is very easy:
1. Install Ollama from the [official website](https://ollama.com)
2. Follow installation guides
3. After installation, select model (llama3 is recommended) and run ollama using command: `ollama run llama3`
4. In macai LLM settings, set ChatGPT/LLM API Url to `http://localhost:11434/api/chat`:
5. In macai New Chat settings, set model to `llama3`
6. Changing default instructions is recommended
7. Test and enjoy!
## Screenshots
### Starting screen
### Customized system message
An example of custom system message and ChatGPT responses:
### Code formatting and syntax highlighting
The syntax of the code provided in ChatGPT response will be highlighted ([185 languages](https://github.com/raspu/Highlightr) supported)
### Table formatting
In most cases, tables in ChatGPT repsonses can be formatted as follows:
### Settings
## License
[MIT](https://github.com/Renset/macai/blob/main/LICENSE.md)