Awesome video. Really helpful and thanks for bringing up the "evals". I never would have known that Agentic frameworks can have a testing mechanism if it wasn't for the "evals" part of this video. Honestly, it would have been even more beneficial to the viewers if you had run few commands testing the examples.
Thanks for making this. Good video and good overview. I’ve tried crewAI and Langchain and I’m not a fan because they’re extremely complex for no reason and under the hood all they provide is an API wrapper on top of OpenAI. Could you try using this with an open source LLM server, like Ollama? That would be interesting to see. The simplicity of this library is very welcome, but the need to use OpenAIs API is a non starter for many. We can’t productionize this because that means our production becomes dependent on OpenAI’s latency, pricing, and restrictions
Valid point. The point of the video (and the library) is to show the repo and how people can just build their own framework instead of using some libraries like LanGraph or Crew.ai. Also that library is much easier to use than these but is not really intended to use in production (yet?) I will do a live building session next week in my Skool community to show how to use it and build an agent live
swarm is counter-intuitive to AI. The AI is suppose to build, grow, refine, manage logical structures all on its own("behind the scenes") just like its learning and growing just by users inputs(prompts). Users aren't suppose to take this capability from AI, its suppose to do this in its "head" all while being prompted...probably very little productivity gains in actuality with swarm.
Hype vid. Other frameworks been out awhile.
Awesome video. Really helpful and thanks for bringing up the "evals". I never would have known that Agentic frameworks can have a testing mechanism if it wasn't for the "evals" part of this video. Honestly, it would have been even more beneficial to the viewers if you had run few commands testing the examples.
Thanks for the feedback. I will do that next time ;)
Thanks for making this. Good video and good overview. I’ve tried crewAI and Langchain and I’m not a fan because they’re extremely complex for no reason and under the hood all they provide is an API wrapper on top of OpenAI.
Could you try using this with an open source LLM server, like Ollama? That would be interesting to see. The simplicity of this library is very welcome, but the need to use OpenAIs API is a non starter for many. We can’t productionize this because that means our production becomes dependent on OpenAI’s latency, pricing, and restrictions
Why didnt you deploy or try an agent?
Because you can just do it yourself
So can anyone read the information available on public domain
Valid point. The point of the video (and the library) is to show the repo and how people can just build their own framework instead of using some libraries like LanGraph or Crew.ai. Also that library is much easier to use than these but is not really intended to use in production (yet?)
I will do a live building session next week in my Skool community to show how to use it and build an agent live
@@DonaAI Thank you
swarm is counter-intuitive to AI. The AI is suppose to build, grow, refine, manage logical structures all on its own("behind the scenes") just like its learning and growing just by users inputs(prompts). Users aren't suppose to take this capability from AI, its suppose to do this in its "head" all while being prompted...probably very little productivity gains in actuality with swarm.
you just read the repo lol, use it to test it or build some demos... bad video imo
Accent Français
Difficile d'y échapper 🤭
They copied Agent Swarms and Crewai
And they copied someone else.