crewAI
crewAI copied to clipboard
Framework for orchestrating role-playing, autonomous AI agents. By fostering collaborative intelligence, CrewAI empowers agents to work together seamlessly, tackling complex tasks.
Problem: Crew runs correctly to completion, then starts over with new "made up" input. Expected: When Crew has completed job, program ends. Started with a version of email classifier demo...
from crewai import Agent, Task, Crew from langchain_openai import ChatOpenAI import os os.environ["OPENAI_API_KEY"] = "NA" llm = ChatOpenAI( model = "crewai-llama2", base_url = "http://localhost:11434/v1") general_agent = Agent(role = "Math Professor",...
fix to the import statement in PGSearchTool documentation
This PR has the intention to add mypy as the type-checker and also update the GH actions to reflect that. Along with that, some fixes were done on the typing,...
Very odd one here, this tool works like a dream with GPT4, however when I moved to deepseek cloud api (muuuuuch cheaper but looks awesome), the crew execution chain or...
CrewAI has a dependency on top of embedchain. Latest embedchain (0.1.103) requires ChromaDB with version >= 0.5.0 while crewai-tools depends on versions
More like a KT/Question. I am trying to have RAG support for Open LLM (mamba), But unable to get it to work. I am following this video, but Code is...
I using Crew AI 0.28.8. i used the callback function according to crew AI documentation but now the callback function is not working on the crew AI. def callback_function(output: TaskOutput):...
from langchain_community.llms import Ollama from crewai import Agent, Crew, Task, Process model = Ollama(model = "ollama run llama3:8b") Itinerary = "vacation through the colorado and the surrounding rocky mountains" Travel_Agent...
I tried to run the sample code using Azure OpenAI API from [https://github.com/joaomdmoura/crewAI-examples/blob/main/azure_model/main.py](url) I haved made one fix for the original code to include the `expected_output='Detailed report on potential AI...