Jiva Framework icon
Jiva Framework icon

Jiva Framework

An Agent framework powered by LLMs with experience of time, autonomous execution, short term and long term memory.

Cost / License

  • Free
  • Open Source

Platforms

  • Self-Hosted
  • Docker
-
No reviews
0likes
0comments
0news articles

Features

Suggest and vote on features
  1.  AI-Powered

 Tags

  • autonomous-agents
  • mistral-ai
  • Ollama

Jiva Framework News & Activities

Highlights All activities

Recent activities

Show all activities

Jiva Framework information

  • Developed by

    Karmaloop AI
  • Licensing

    Open Source (MIT) and Free product.
  • Written in

  • Alternatives

    12 alternatives listed
  • Supported Languages

    • English

AlternativeTo Category

AI Tools & Services

GitHub repository

  •  19 Stars
  •  5 Forks
  •  4 Open Issues
  •   Updated  
View on GitHub

Popular alternatives

View all
Jiva Framework was added to AlternativeTo by Abi-Dev on and this page was last updated .
No comments or reviews, maybe you want to be first?
Post comment/review

What is Jiva Framework?

An AI Agent with Temporal Awareness and Ethical Decision-Making

Jiva Framework is an innovative open-source project aimed at creating an AI agent that experiences time, forms memories, and operates based on ethical principles. This framework provides a unique approach to AI development, incorporating concepts such as cyclical time perception, ethical decision-making, and continuous learning.

Key Features:

  • Temporal Awareness: Jiva operates on a day/night cycle, allowing for a more human-like perception of time.
  • Memory Systems: Utilizes both short-term and long-term memory, powered by vector databases for efficient storage and retrieval.
  • Ethical Framework: Incorporates ethical principles into decision-making processes, ensuring responsible AI behavior.
  • Task Management: Autonomously generates, prioritizes, and executes tasks to achieve given goals.
  • Adaptive Learning: Engages in cyclical learning and refinement of knowledge through regular "sleep" cycles. There is more to come on this, see below.
  • Sensor Integration: Modular design allows for easy integration of various input sensors. Supports only human-input as of now.
  • Action Management: Actions registry is meant to continually grow and become richer as development continues.

Long-term goals for Adaptive Learning A key goal of the project is to allow for the agent to sleep and fine-tune its underlying LLModel by consolidating its thoughts and actions throughout the day.

Official Links