Ollama AI

I did find a video on Youtube showing the installation and usage of Ollama AI. The Ollama AI runs on your local machine. It is currently available for Linux and macOS. And runs in the CLI. You can install multiple language models such as: llama2 and zephyr.

Here is an example with the zephyr language model:

$ ollama run zephyr
>>> How can I write a "Hello world!" program in Go?
To write a "Hello, world!" program in Go, follow these steps:

1. Open your preferred text editor or code editor and create a new file.
2. Save the file with a `.go` extension (e.g., `hello.go`).
3. Copy and paste the following code into your new file:

package main

import "fmt"

func main() {
    fmt.Println("Hello, world!")
}
  1. Save the file and open a command prompt or terminal window in the directory where the hello.go file is saved.

  2. Compile your program using the Go compiler by running the following command:

    go build hello.go

  3. Once the program has finished compiling, run it by executing the following command:

    ./hello
  4. The output of your program should be Hello, world!. Enjoy your first Go program!



So this is the answer by Ollama zephyr.

There is also an Emacs plugin: Ellama. I tried it with Doom Emacs (Emacs 28.2). After the installation the Ellama plugin works. But if I close Emacs and start Emacs again, then the Ellama plugin is not available in the Menu. I don’t know what is going on there! Maybe there is some compatibility issue with Doom Emacs?