Non-GPU

At a glance

The community member is trying to use the example from the LlamaIndex documentation with the ollama library, but the process of indexing a simple text file is taking more than 4 minutes to complete. They are using a cloud-based machine with 32 GB RAM and 4 core CPU. The community member is wondering if there is a way to speed up the process and also notes that the documents still use LLaMA 3, suggesting an update to 3.1 would be appreciated.

In the comments, another community member suggests that non-GPU machines tend to take more time compared to GPU-enabled machines.

Useful resources

SSumitPandit

I am trying this example https://docs.llamaindex.ai/en/stable/getting_started/starter_example_local/ with ollama, now to query simple file, it's taking more than a 4 minutes to respond and complete the script.
I am using this text file https://sherlock-holm.es/stories/plain-text/advs.txt
It looks like this step is taking lot of time

Plain Text

index = VectorStoreIndex.from_documents(
    documents,
)

I am using 32 GB RAM with 4 core CPU in cloud.
Is there anyway I can speed up the process.
Also, I see the documents still use llama3, would be great if updated to 3.1

Attachment

1 comment

WWhiteFang_Jr

Yea Non-GPU machines tend to take more time in comparison to GPU ones

Add a reply

Find answers from the community

Non-GPU