pavel-one/EdgeGPT-Go

🪄 EdgeGPT-Go

Feature:

• GRPC interface
• Library interface
• Sync/Async request
• CLI interface
• Refresh session

How to use it:

Getting authentication (Required)

• Install and open Microsoft Edge browser
• Install Cookie-Editor extension
• Go to New Bing login and getting access
• Open the extension
• Click "Export" on the bottom right, then "Export as JSON" (This saves your cookies to clipboard)
• Create folder cookies
• Paste your cookies into a file 1.json

If you have several accounts - repeat for each of them and save to the cookies folder

Use as a chat cli app

docker run -it -v ./cookies:/app/cookies ghcr.io/pavel-one/edgegpt-chat

Or download binary in release page

Use as a library

go get github.com/pavel-one/EdgeGPT-Go

package main

import (
	"github.com/pavel-one/EdgeGPT-Go"
	"log"
)

func main() {
	s := EdgeGPT.NewStorage()

	gpt, err := s.GetOrSet("any-key")
	if err != nil {
		log.Fatalln(err)
	}

	// send ask async
	mw, err := gpt.AskAsync("Hi, you're alive?")
	if err != nil {
		log.Fatalln(err)
	}

	go mw.Worker() // start worker

	for range mw.Chan {
		// update answer
		log.Println(mw.Answer.GetAnswer())
		log.Println(mw.Answer.GetType())
		log.Println(mw.Answer.GetSuggestions())
		log.Println(mw.Answer.GetMaxUnit())
		log.Println(mw.Answer.GetUserUnit())
	}

	// send sync ask
	as, err := gpt.AskSync("Show an example of sockets in golang gorilla")
	if err != nil {
		log.Fatalln(err)
	}

	log.Println(as.Answer.GetAnswer())
}

Use as a docker microservice

docker run -v ./cookies:/app/cookies -p 8080:8080 ghcr.io/pavel-one/edgegpt-grpc:latest

Use as a docker-compose

version: "3"
services:
  gpt:
    image: ghcr.io/pavel-one/edgegpt-grpc:latest
    restart: unless-stopped
    ports:
      - "8080:8080"
    volumes:
      - ./cookies:/app/cookies

Available environment

Example service

Work progress...

FAQ:

Change protoc

If you change protoc file, use protoc --go_out=. --go-grpc_out=. proto/gpt.proto

Stopped after 10 redirects

If you use this library in China set environment:

CONVERSATION_ENDPOINT=https://edge.churchless.tech/edgesvc/turing/conversation/create

Example:

docker run -e CONVERSATION_ENDPOINT=https://edge.churchless.tech/edgesvc/turing/conversation/create -v ./cookies:/app/cookies -p 8080:8080 ghcr.io/pavel-one/edgegpt-grpc:latest

Or docker-compose:

version: "3"
services:
  gpt:
    image: ghcr.io/pavel-one/edgegpt-grpc:latest
    restart: unless-stopped
    environment:
      - CONVERSATION_ENDPOINT=https://edge.churchless.tech/edgesvc/turing/conversation/create
    ports:
      - "8080:8080"
    volumes:
      - ./cookies:/app/cookies