MetaGPT: Redefining Multi-Agent Collaboration for Complex Tasks
ฝัง
- เผยแพร่เมื่อ 2 มิ.ย. 2024
- ►Subscribe to my Newsletter (My AI updates and news clearly explained): louisbouchard.substack.com/
References:
►Read the full article: www.louisbouchard.ai/metagpt/
►Hong et al., 2023: MetaGPT, arxiv.org/pdf/2308.00352.pdf
►Code: github.com/geekan/MetaGPT/blo...
►Twitter: / whats_ai
►Support me on Patreon: / whatsai
►Join Our AI Discord: / discord
How to start in AI/ML - A Complete Guide:
►www.louisbouchard.ai/learnai/
Become a member of the TH-cam community, support my work and get a cool Discord role :
/ @whatsai
#GPT #ChatGPT #MetaGPT - วิทยาศาสตร์และเทคโนโลยี
Very Well Explained. Thank You
Lmao we went from replacing workers with gpt to replacing whole companies , impressive feat to say the least.
Is part of the problem perhaps that we just aren't training these LLMs for the concept of IDK? Some questions don't have answers yet, yet we train the models to assume there always is one.
I keep waiting for the day when the CGPT response is "No effing clue!", but that's likely a long ways off. It would be a much more acceptable response than sending me chasing after imaginary research for example.
As for agents, extremely cool, though I'm very surprised at how quickly they have advanced. Seems it wasn't that long ago someone first hooked up CGPT to itself, but then lately I've been mired in the land of Stable Diffusion.... nuff said.
I am not a robot (though I might be a NaN).
🖖😎👍
Good question. LLM-based solutions are often easy to demo but hard to finalize in production.
Great explanation video
Super interesting, thanks for sharing!
mutl-agent collaboration framework is the key here
has any one experiences with the latency when multiple agents interact?
❤
or, have just a better model that does not allucinate
Haha simple enough!