Join Our Community for Enhancing Open-Source LLM Integration and More
ElliottDyson opened this issue · comments
Hello Fellow Developers and Original Authors 👋,
Firstly, I'd like to express my immense appreciation for the work done by Joon Sung Park, Joseph C. O'Brien, Carrie J. Cai, Meredith Ringel Morris, Percy Liang, and Michael S. Bernstein. Your contributions to the field, as showcased in this repository (and your original paper), are groundbreaking and serve as a solid foundation for further explorations.
As I admire your work, I've been working on adapting this remarkable codebase to integrate with open-source LLMs, like LLAMA/LLAMA-2. While navigating the issues page, I discovered that others are also interested in contributing to this project along similar lines.
This led me to consider forming a community focused on:
- Improving Performance: Further optimising these agents based on your foundational work (testing prompting methods/techniques, parallelisation of agents, etc.,).
- Generalisation: Aiming for greater adaptability to facilitate the project's integration into diverse applications.
- Open-Source LLM Compatibility: Ensuring the project's seamless operation with open-source LLM frameworks.
For those interested in participating, here's how to get involved:
- Discord Community: Join the conversation here.
- GitHub Repo: Find the repository here
If you share the enthusiasm for building upon this excellent foundation, please feel free to join us. Together, we can push the boundaries even further, standing on the shoulders of giants.
Best regards,
Elliott Dyson
Citation: For anyone joining this endeavour, kindly remember to cite the original paper when leveraging the code or data:
```
@inproceedings{Park2023GenerativeAgents,
author = {Park, Joon Sung and O'Brien, Joseph C. and Cai, Carrie J. and Morris, Meredith Ringel and Liang, Percy and Bernstein, Michael S.},
title = {Generative Agents: Interactive Simulacra of Human Behavior},
year = {2023},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
booktitle = {In the 36th Annual ACM Symposium on User Interface Software and Technology (UIST '23)},
keywords = {Human-AI interaction, agents, generative AI, large language models},
location = {San Francisco, CA, USA},
series = {UIST '23}
}
```
Interesting! Actually I have modified a version for my own research on my Mac M1. It is using llama.cpp as the local LLM and is running quite smoothly. Glad someone has an initiative to make this more generic and open. I will see how I can contribute.
Interesting! Actually I have modified a version for my own research on my Mac M1. It is using llama.cpp as the local LLM and is running quite smoothly. Glad someone has an initiative to make this more generic and open. I will see how I can contribute.
Would love to see you on the discord!
Interesting! Actually I have modified a version for my own research on my Mac M1. It is using llama.cpp as the local LLM and is running quite smoothly...
can you explain what exactly you modified? I also want to run on llama.cpp
Interesting! Actually I have modified a version for my own research on my Mac M1. It is using llama.cpp as the local LLM and is running quite smoothly...
can you explain what exactly you modified? I also want to run on llama.cpp
There's too much to cover in a simple message I'm afraid, you're welcome to look at the GitHub repo.
Interesting! Actually I have modified a version for my own research on my Mac M1. It is using llama.cpp as the local LLM and is running quite smoothly. Glad someone has an initiative to make this more generic and open. I will see how I can contribute.
Hi, what model are you using?