OS-Copilot

@oscopilot

The library for generalist computer agents research and applications. Github: github.com/OS-Copilot/OS-… Discord: discord.gg/rXS2XbgfaD

Anywhere On Earth

os-copilot.github.io

Replies Reposts Media

OS-Copilot
@oscopilot
Nov 4, 2024
We have a new member on board! 😎 Who's next?
Zhiyong Wu
@zywu_hku
Nov 4, 2024
🚀 Excited to introduce a new member of the OS-Copilot family: OS-Atlas - a foundational action model for GUI agents Paper: huggingface.co/papers/2410.23… Website: osatlas.github.io A thread on why this matters for the future of OS automation 🧵 TL;DR: OS-Atlas offers: 1.
OS-Copilot
@oscopilot
May 23, 2024
🚀 New release v0.1.2 1⃣ Optimized Continuous Self-Learning 2⃣ Lightweight Friday. Just 100 lines of code to build a strong lightweight OS agent, easy for further extension. 3⃣ New Fault Tolerance Mechanism Tagging to share your demos! or @zywu_hku
OS-Copilot
@oscopilot
May 14, 2024
Thanks for posting! We definitely learned a lot from @langchain during the early stage of OS-Copilot's development. And several exciting new features are coming this week. PRs and contributions are highly welcome!
LangChain
@LangChain
May 11, 2024
🤖OS-Copilot ⚡️Towards Generalist Computer Agents with Self-Improvement⚡️ Fantastic open-source repo for building tool-using agents to automate our daily tasks Paper: arxiv.org/pdf/2402.07456 GitHub: github.com/OS-Copilot/OS-…
1
OS-Copilot
@oscopilot
Apr 3, 2024
Yeah. We are working on this new feature (multi-agent cooperation). Contributing is welcome!😃
dinos
@din0s_
Mar 11, 2024
Replying to @din0s_
OS-Copilot is an interesting first step in this direction, which previously held the top position in the GAIA benchmark: arxiv.org/abs/2402.07456 x.com/din0s_/status/…
OS-Copilot
@oscopilot
Apr 3, 2024
We have uploaded a one-click SDL script to our latest release. Come and play with the new code base
Arghya Bhattacharya
@orgho12
Mar 6, 2024
I’ve been working on a toy implementation of the OS-copilot based FRIDAY framework. There is a significant improvement in model’s Self-Directed Learning capabilities using Opus, I’m a 100% sure if i run the GAIA benchmarks on this model, It’s going to shatter level-3 tasks scores