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.
I've added commented out code that can be used to use local model instead of OpenAI. (It works!)
please see the screenshot below. running ollama directly - not problem, but running crewais.py yields: name 'ollama' is not defined ![ollama](https://github.com/joaomdmoura/crewAI/assets/69711075/d890d4ef-10c7-4d9e-ad0f-9838cc5150b9)
For folks not trying to have each agent use a different LLM this will allow them to set it once on the crew and forget about it
When using CrewAI with the Free Gemini Pro API that has a limit of 60 API calls per minute. It would be incredibly helpful to have a simple way to...
One can easily use human input as tool by using the langchain tool but would be nice to add proper docs around it as it seem a usual use case...
I have been trying the example on the landing page using Ollama and TextGen-Webui's OpenAI API wrapper. I followed all the steps, but the writer doesn't write the summary at...
Posted this originally on the examples repo but more visibility here and others may run into same issue. I run the web search example using an open-source model via a...
Currently, the system seems to only support text-based agents, would be nice to add support for multimodal models. Autogen's implementation: https://github.com/microsoft/autogen/blob/main/autogen/agentchat/contrib/multimodal_conversable_agent.py It seems non-straightforward to make the langchain memory modules...
A feature request for the wiki 1. A section on configuring CrewAI with different LLM providers, as well as local model providers such as LM-Studio, llama.cpp etc. So far, I...
Notebook showing how to use Ollama