parakeet

README

🦜 Parakeet

Parakeet is the simplest Go library to create GenAI apps with Ollma.

A GenAI app is an application that uses generative AI technology. Generative AI can create new text, images, or other content based on what it's been trained on. So a GenAI app could help you write a poem, design a logo, or even compose a song! These are still under development, but they have the potential to be creative tools for many purposes. - Gemini

✋ Parakeet is only for creating GenAI apps generating text (not image, music,...).

Install

go get github.com/parakeet-nest/parakeet

Simple completion

The simple completion can be used to generate a response for a given prompt with a provided model.

package main

import (
	"github.com/parakeet-nest/parakeet/completion"
	"github.com/parakeet-nest/parakeet/llm"
	"fmt"
	"log"
)

func main() {
	ollamaUrl := "http://localhost:11434"
	model := "tinydolphin"

	options := llm.Options{
		Temperature: 0.5,  // default (0.8)
	}

	question := llm.Query{
		Model: model,
		Prompt: "Who is James T Kirk?",
		Options: options,
	}

	answer, err := completion.Generate(ollamaUrl, question)
	if err != nil {
		log.Fatal("😡:", err)
	}
	fmt.Println(answer.Response)
}

Simple completion with stream

package main

import (
	"github.com/parakeet-nest/parakeet/completion"
	"github.com/parakeet-nest/parakeet/llm"
	"fmt"
	"log"
)

func main() {
	ollamaUrl := "http://localhost:11434"
	model := "tinydolphin"

	options := llm.Options{
		Temperature: 0.5, // default (0.8)
	}

	question := llm.Query{
		Model: model,
		Prompt: "Who is James T Kirk?",
		Options: options,
	}
	
	answer, err := completion.GenerateStream(ollamaUrl, question,
		func(answer llm.Answer) error {
			fmt.Print(answer.Response)
			return nil
		})

	if err != nil {
		log.Fatal("😡:", err)
	}
}

Completion with context

see: https://github.com/ollama/ollama/blob/main/docs/api.md#generate-a-completion

The context can be used to keep a short conversational memory for the next completion.

package main

import (
	"github.com/parakeet-nest/parakeet/completion"
	"github.com/parakeet-nest/parakeet/llm"

	"fmt"
	"log"
)

func main() {
	ollamaUrl := "http://localhost:11434"
	model := "tinydolphin"

	options := llm.Options{
		Temperature: 0.5, // default (0.8)
	}

	firstQuestion := llm.Query{
		Model: model,
		Prompt: "Who is James T Kirk?",
		Options: options,
	}

	answer, err := completion.Generate(ollamaUrl, firstQuestion)
	if err != nil {
		log.Fatal("😡:", err)
	}
	fmt.Println(answer.Response)

	fmt.Println()

	secondQuestion := llm.Query{
		Model: model,
		Prompt: "Who is his best friend?",
		Context: answer.Context,
		Options: options,
	}

	answer, err = completion.Generate(ollamaUrl, secondQuestion)
	if err != nil {
		log.Fatal("😡:", err)
	}
	fmt.Println(answer.Response)
}

Chat completion

The chat completion can be used to generate a conversational response for a given set of messages with a provided model.

package main

import (
	"github.com/parakeet-nest/parakeet/completion"
	"github.com/parakeet-nest/parakeet/llm"

	"fmt"
	"log"
)

func main() {
	ollamaUrl := "http://localhost:11434"
	model := "deepseek-coder"

	systemContent := `You are an expert in computer programming.
	Please make friendly answer for the noobs.
	Add source code examples if you can.`

	userContent := `I need a clear explanation regarding the following question:
	Can you create a "hello world" program in Golang?
	And, please, be structured with bullet points`

	options := llm.Options{
		Temperature: 0.5, // default (0.8)
		RepeatLastN: 2, // default (64)
		RepeatPenalty: 2.0, // default (1.1)
	}

	query := llm.Query{
		Model: model,
		Messages: []llm.Message{
			{Role: "system", Content: systemContent},
			{Role: "user", Content: userContent},
		},
		Options: options,
		Stream: false,
	}

	answer, err := completion.Chat(ollamaUrl, query)
	if err != nil {
		log.Fatal("😡:", err)
	}
	fmt.Println(answer.Message.Content)
}

✋ To keep a conversational memory for the next chat completion, update the list of messages with the previous question and answer.

I plan to add the support of bbolt in the incoming v0.0.1 of Parakeet to store the conversational memory.

Chat completion with stream

package main

import (
	"fmt"
	"log"

	"github.com/parakeet-nest/parakeet/completion"
	"github.com/parakeet-nest/parakeet/llm"
)

func main() {
	ollamaUrl := "http://localhost:11434"
	model := "deepseek-coder"

	systemContent := `You are an expert in computer programming.
	Please make friendly answer for the noobs.
	Add source code examples if you can.`

	userContent := `I need a clear explanation regarding the following question:
	Can you create a "hello world" program in Golang?
	And, please, be structured with bullet points`

	options := llm.Options{
		Temperature: 0.5, // default (0.8)
		RepeatLastN: 2, // default (64) 
	}

	query := llm.Query{
		Model: model,
		Messages: []llm.Message{
			{Role: "system", Content: systemContent},
			{Role: "user", Content: userContent},
		},
		Options: options,
		Stream:  false,
	}

	_, err := completion.ChatStream(ollamaUrl, query,
		func(answer llm.Answer) error {
			fmt.Print(answer.Message.Content)
			return nil
		})

	if err != nil {
		log.Fatal("😡:", err)
	}
}

Documentation

Index

• Variables

Variables

var About = "🦜 Parakeet v0.0.0 🪺 [nest]"

var Version = "v0.0.0"