CrewAi + Solar/Hermes + Langchain + Ollama = Super Ai Agent (Fully Local)

#datascience #artificialintelligence #automation #llm #deeplearning
In this video, you'll learn what is CrewAi, architecture design, the differences between Autogen, ChatDev, and Crew Ai, and how to use Crew Ai, Langchain, and Solar or Hermes Power by Ollama to build a super Ai Agent

Timestamps:
0:00 - what you will learn
0:45 - What is Crewai
1:21 - a key feature
2:09 - Autogen Vs AChatdev Vs Crewai
3:08 - install ollama
3:52 - let's start coding
06:46 - Conclusion

FOLLOW ME :

[Must Read]

[Must Watch]

[Link Used]

Additional Tags and Keywords:
#AI #Chatbot #AutoGen #chatdev #Crewai #FunctionCall #SuperAI #python #Technology #Innovation #ArtificialIntelligence #ollama #MachineLearning #NaturalLanguageProcessing #VirtualAssistant #Automation #HumanComputerInteraction #ConversationalAI #openhermes #solar #Tech #video #llmlingua #microsoftresearch

[GaoDalie_AI]

Thank you so much for watching guys! I would highly appreciate it if you


FOLLOW ME :

[justinzhang]

THX for introducing crewAI. That seems much easier now.

[olucasfranco]

I liked the video, now a complete tutorial on how to set up a team just like you showed in the video, it would be even cooler.

[HammerOnTheNet]

Really really cool!
Waiting for next videos, great job!!!

[jbo]

Hallucinated output is pretty normal for open source llms. Basic tasks can be done with a lot of prompt eng work, but more advanced functions will take better models or a lot of work under the hood 🤔

[IdPreferNot]

Really appreciate the concise, yet thoughtful run throughs. Once again looking forward to more of your videos. With these team apps, an interesting video would be to look at tools for teams building teams.... ie to help figure out what teams, tasks, etc might be helpful to approach various larger tasks. Another would be listing sources of some of the best data, process, etc APIs, and how to teach these agents to use. For example... can the human get an API key from some random app or webservice (eg newspaper, and somehow teach the crewai agent how to access it with your key and how to utilize its functionality?

[frankbradford]

FYI, This does not work. error . It looks good but does work at all. Do you have a solution?
main.py", line 8, in <module>
ollama_openhermes = ollama(model='openhermes')

TypeError: 'module' object is not callable

ERROR: No matching distribution found for Jinja>=3.1.2

[themaxgo]

FYI: does not work (anymore), getting a TypeError: 'module' object is not callable @ ollama_openhermes = ollama(model="openhermes" (same for "solar).
fix: changing to "langchain_community.llms import Ollama (uppercase O) and changing the call to define the 2 models works...

[scitechtalktv]

Very interesting would be the ability for function calling and tool use for open source LLMs. It appears that many people have difficulty getting that to really working in CrewAI applications they have tried! Which open source LLMs can really do function calling and use tools (such as calculator, webscraping, Internet searching, searching Arxiv articles, searching PubMed articles, etc etc). Can you make YouTube videos about successfully doing those things? Would be very welcome!

[cyborgmetropolis]

Had me going until “Crew AI” was pronounced “creweye”