Ollama Status Code 404 Error when trying to run with LangChain

Question

I am getting the following error when trying to run Ollama with LLama3 and invoking the model from LangChain (python)

langchain_community.llms.ollama.OllamaEndpointNotFoundError: Ollama call failed with status code 404. Maybe your model is not found and you should pull the model with `ollama pull llama3`.

Context:

• Running on MacOs
• Code:

from langchain_community.llms import Ollama

llm = Ollama(model="llama3", base_url="http://localhost:11434/")
llm.invoke("Why is the sky blue?")

• Tried running Ollama as a service with ollama serve (It does not seem to make a difference)

• I am able to see that Ollama is running on the localhost:11434

• I am getting a 404 error when I try to access localhost:11434/llama3

• ollama list shows llama 3 installed

Answer 1

I did faced the same issue with both llama3 and llama2 on my Mac

Here is how it got resolved-

1. Remove the ending slash from the base URL param

llm = Ollama(model="llama3", base_url="http://localhost:11434")

2. Restart the Jupytar karnel.
3. Run all the cells again.

Answer 2

It works for me in Jupyter notebook when I set model="llama2"

llm = Ollama(model="llama2")