Skip to main content
Pre-release

This release is a pre-release and may not be stable for production use.

AI-Completer

A totally automatic AI to interact with environments.

Now it only supports OpenAI API.

Setup

Note:Please use stable branch if the dev branch is broken

  1. Clone the Repositry
git clone https://github.com/AI-Completer/AI-Completer.git
cd AI-Completer
  1. Modify the config file

You can use either vim or your editor to modify the file config-example.json, and save it to config.json

  1. Run and enjoy yourself

Note: Use Python 3.11+ to run this program.

python3 -m aicompleter talk --ai bingai
# This will start a conversation with bing ai.
python3 -m aicompleter helper --enable-agent --include pythoncode
# This will start a helper with OpenAI API

You can add custom interface to this program to add more function.

Quick Start

The code below will start a talk with gpt, the detailed model varies with your configure, default is gpt-3.5-turbo.

from aicompleter import *
from aicompleter.implements import *
import asyncio

cfg = Config({
    # 'openaichat' is the namespace, you can change to your own namespace
    'openaichat':{
        'openai':{
            'api-key': 'sk-...',                 # Put you OpenAI API key here
            'model': 'gpt-3.5-turbo',            # AI model, default is gpt-3.5-turbo
        }
    }
})
# ChatAI, use openai model gpt-3.5-turbo-0301
chater = ai.openai.Chater(cfg['openaichat'])
# Chat Interface, based on chater -> OpenAI API
chatinterface:ai.ChatInterface = ai.ChatInterface(ai=chater, namespace='openaichat')
# Handler, interacting between interfaces
hand:Handler = Handler(cfg)

async def main():
    # Add Interfaces
    await hand.add_interface(chatinterface)
    # Start a new session
    session:Session = await hand.new_session()
    print("Please start a conversation")
    while True:
        # Get input from console
        word = input(">>> ")
        if word == 'exit':
            break
        # Send a ask command to the chat interface, the ai is asked by the content (text, the question of user)
        print(await session.asend('ask', word))

# Start the loop
asyncio.run(main())

Examples

See examples

Document

Reference: Document

To-do List

We are adding more support to this program.

  • Add Commands Intergation with AI model.
    • Add Commands Support
  • Add memory system
    • Add History
    • Add Memory Analyse
  • Add Extra Interface Package

Metadata

Release files for aicompleter 0.0.1rc5

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for aicompleter 0.0.1rc5
File Size Uploaded
aicompleter-0.0.1rc5.tar.gz 104.7 kB Details

Built distribution (wheel)

Table of built distributions (wheels) for aicompleter 0.0.1rc5
File Interpreter ABI Platform
aicompleter-0.0.1rc5-py3-none-any.whl Python 3 none any Details

Total release size: 231.8 kB

Release files / aicompleter-0.0.1rc5.tar.gz

Download URL aicompleter-0.0.1rc5.tar.gz
Size 104.7 kB
Tags Source
SHA-256 checksum
How to use checksums
9782e5a6429e33b0af15838b43b9416d3f8b30c8d1b51fd466aa1a0e30120d19
BLAKE2b-256 checksum
How to use checksums
a400592c3ae2f972ac2f3aef59e25bbfcae79a35501b06fbf9e7abcc0804790a
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/4.0.2 CPython/3.11.1

Release files / aicompleter-0.0.1rc5-py3-none-any.whl

Download URL aicompleter-0.0.1rc5-py3-none-any.whl
Size 127.1 kB
Tags Python 3
SHA-256 checksum
How to use checksums
7e67b248cc50d49e32439a7c6f17c37f600638b7f4bfa9be71ce997a42a67288
BLAKE2b-256 checksum
How to use checksums
17c4347268c08207713b081015cf5005c448ce701fd96efb62b5cccb08e9b158
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/4.0.2 CPython/3.11.1

Release history Release notifications | RSS feed

This release

0.0.1rc5 This release

2 release files

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page