/Nuggt

An Autonomous LLM Agent that runs on Wizcoder-15B

Primary LanguagePythonCreative Commons Zero v1.0 UniversalCC0-1.0

Nuggt: An Autonomous LLM Agent that runs on Wizcoder-15B (4-bit Quantised)

GitHub Repo stars Twitter Follow Discord Follow

This Repo is all about democratising LLM Agents with powerful Open Source LLM Models. Join us on this exciting journey of task automation with Nuggt, as we push the boundaries of what can be achieved with smaller open-source large language models, one step at a time 😁.

Download Wizcoder-15B 4-bit Quantised from: https://huggingface.co/TheBloke/WizardCoder-15B-1.0-GPTQ (PS: TheBloke is a legend)

Demo: RandomForest Implementation on Titanic Dataset

ML-Fast.mp4

Features

  • 🗃️ All-in-one platform: Create any functionality using a single set of instructions.
  • 🧠 Powered by Wizcoder-15B 4-bit Quantised
  • 🔗 Build Controlled Bots: Develop controlled autonomous bots capable of automating tasks efficiently.
  • 🌐 Tool Integration: Enjoy seamless integration with various tools, including Python, document processing, video analysis, Google search, and web browsing.
  • 🚀 Community-Driven: Participate in a vibrant community, contributing feedback, suggestions, and improvements to shape the project's development.

Quickstart

You can setup the API keys either under .env file or set it up under settings tab directly on the Nuggt platform.

git clone https://github.com/Nuggt-dev/Nuggt.git
pip install -r requirements.txt
cd nuggt-wizcoder
Setup model path in server.py
python3 server.py
streamlit run nuggt.py

Documentation

Limitations

This repository serves as an experimental project and should be considered with the following limitations in mind:

  • As of June 9th, 2023, Nuggt is not a fully developed project. We strongly recommend thorough testing before deploying the solution to ensure the security and functionality of your application.

Disclaimer

Please note the following disclaimer before using this open-source project:

  1. No Guarantees: This project is provided as-is, without any warranties or guarantees of any kind. The developers and contributors of this project cannot be held responsible for any direct or indirect damages or issues that may arise from the use of this project.

  2. Experimental Nature: This project is experimental in nature and may not be suitable for production environments or critical systems. It is highly recommended to thoroughly test and evaluate the project before deploying it in any mission-critical or sensitive applications.

  3. Security Considerations: While efforts have been made to ensure the security of this project, it is essential to note that no software is entirely immune to vulnerabilities. Users are responsible for conducting their own security assessments and implementing necessary security measures when utilizing this project.

  4. Third-Party Tools and Services: This project incorporates the use of various third-party tools and services. The developers and contributors do not have control over the availability, functionality, or security of these tools and services. Users are advised to review and comply with the terms, policies, and guidelines of these third-party tools and services when using them in conjunction with this project.

  5. Financial Considerations: Certain tools and services integrated with this project may incur costs, such as cloud computing resources or API usage fees. Users should be aware of the potential financial implications and are responsible for managing and monitoring any expenses incurred while using this project.

  6. User Responsibility: Users of this project are solely responsible for ensuring the legality, appropriateness, and compliance of their prompts, inputs, and actions performed using this project. It is crucial to respect the terms of service, intellectual property rights, and privacy of others when utilizing this project.

By using this project, you acknowledge and agree to the above disclaimer.

Connect with Us