Skip to main content
Pre-release

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

ASIS: Artificial Intelligence Standard Software

Open-Source AI Is Good for Us


Documentation: https://#

Source Code: https://github.com/juanpablo91/Asis


We believe in open source AI to democratize technology, provide opportunities and development, with established standards and rules, creating a transparent and competitive world for the benefit of all.

Asis is a set of standards,rules and framework for the deployment of an ecosystem of AI, based in "Personal AI", Concept analogous to personal computing and how it revolutionized the history of humanity.

Main Features

Some of the main features of ASIS include:

Assistant


Feature Description
Personalized Name, description, model llm, a Basic memory. (voice will be added in a future)
Choose the Model Model llm open source o commercial (More info bellow)
Algorithms create or download different algorithms for managing memory, generating agents and autonomous behaviors, manipulating data for the assistant, etc
Agents Agents pre instated for your assistant (5 agents will be dev in beta version)
Tools Tools pre instated in your assistant, for revolve many problems through agents

Platform for development


Feature Description
Communication: Name, description, model llm, a memory,agents(voice agent will be added in a future release)
Libraries Model llm open source o commercial (More info bellow )

TODO::update this table

Get started

Create a Assistant

What is a Entity AI?

A fundamental concept of ASIS is The AI ​​​​Entity, where each person, company, even space or place can be personified in a digital version, starting from a seed of a foundational model, where each of these is managed independently and personalized.

Basic Caracteristis:

  • Memory long time
  • Nickname
  • Description
  • Model (Model llm)
  • Tools default
  • Agents default

Examples:

  • Michael is a makeup artist and his assistant, called "Mills", is his AI version, managing his clients' schedules, answering questions and calls, connecting with his friends' assistants to go out on weekends, exchanging work with his colleagues.

  • John is the CEO of a start-up and "Bert" is the Ai of the company, managing Okrs and always helping employees, with doubts in his field, connecting with every employee assistant, keeping up with everything new, keeping an eye on cybersecurity, paying attention to employee complaints or problems, from the executive layer down.

  • wally is an Ai, in charge of a residential building, in charge together with an agent of a security company, of making sure that everything is always safe, checking by cameras, sensors, entrances and more and together with the assistants of its residents, notifying arrivals of visitors as news about remodeling or changes in the rules of the building.

More information below

Install:

$ pip install asisai
>>> from  asisai import asis

Models available

Architecture

Assistant Architecture

Assistant Architecture

System tools

These are tools by default:

  • Open stream: Agent specialized for navigating the web, opening and working autonomously in programs.

  • Voice: Personalize your assistant's voice, convert voices, and more.

  • Multimedia generator: Agent specific for generating images and video, and modifying them.

  • Asis Chat: Agent specific for chatting between assistants, exchanging information about patterns, schedules, and more, through a friend request.

More tools will be added in the future

Memory

The memory of assistants will be managed by different algorithms for long-term and short-term storage of basic information.

Images, videos, and more complicated algorithms will be added in the future.

🧠 Super Agent

The super agent is the main agent. It receives all the information from its environment, handles decisions, and communicates with the main model to solve complex problems in an autonomous and self-improving way.

Agents

An intelligent agent is an entity capable of perceiving its environment, processing such perceptions, and responding or acting in said environment in a rational manner to achieve objectives.

We are interested in applying code from AutoGPT, which is an open-source project for creating agents.

Disclaimer

Contribution and Roadmap


Join our community

Discord

License

Apache License 2.0

Metadata

Release files for asisai 0.2.1a1

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

Built distribution (wheel)

Table of built distributions (wheels) for asisai 0.2.1a1
File Interpreter ABI Platform
asisai-0.2.1a1-py3-none-any.whl Python 3 none any Details

Release files / asisai-0.2.1a1-py3-none-any.whl

Download URL asisai-0.2.1a1-py3-none-any.whl
Size 7.8 kB
Tags Python 3
SHA-256 checksum
How to use checksums
57f0844e4a99959917d17c8e012517f40b875554eb8e2f7c4b37ade88321df3d
BLAKE2b-256 checksum
How to use checksums
47ce933768df96e32b78a9558edf7d57dd0175b558f272b69737a0bdeec2df98
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/5.0.0 CPython/3.10.4
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