Entity Extraction stuck on 0% with only CPU

[podral3]

After generating embeddings program is stuck.
Extracting entities from chunks: 0%| | 0/42 [00:00<?, ?chunk/s

I'm running lightrag_ollama_demo.py. Tried to run this on multipe llama models but every time it gets stuck on this moment.
I'm running this on Ubuntu 24.04 with: Intel Xeon Gold 6326 @ 2.90GHz

[brandonbevans]

I'm seeing the same running an A100, a pretty strong GPU, on Google Colab with a modified qwen2.5.
There are still requests being served according to the ollama server output, and it looks like it will eventually work. But the progress bar is misleading. It seems like there's something inefficient going on because everything else is blazing fast.