Hello guys~ I need a help,,

Hi, we need more details.
How big is the index? Which model are you using? Which retriever? Which Query Engine?

Are you running your llm on a CPU based machine

Thank you,
The model used is meta-llama/Meta-Llama-3-8B-Instruct,
"The retriever used is based on the ChromaVectorStore,
The query engine used is part of the LlamaIndex framework, which includes a response synthesizer configured with ResponseMode.TREE_SUMMARIZE. The query engine is created from a VectorStoreIndex,
Index is not too big, about 17 PDF, 10 TXT

Oh i thought that i am using my GPU
but, I lose my authority to access to the GPU.
and i solved. Thank you guys for helping me.

Have a good day

you helped yourself 🙂