# Advice on models for RAG use case — retained record

> Retained reddit record from reddit:LocalLLaMA.

Canonical: https://highsignal.app/data/records/18163a6fe8075e1b

[← sources](https://highsignal.app/data)


source: [reddit](https://highsignal.app/data/reddit) (reddit:LocalLLaMA)  published 2026-09-27 09:11 ingested 2026-09-28 02:41

[original source ↗](https://www.reddit.com/r/LocalLLaMA/comments/1wrfjnc/advice_on_models_for_rag_use_case/)

## retained text



```
Hi all,

I'm looking for some advice picking models. I'm looking to try a project to ingest quite a large amount of docments into a knowledgebase, allowing me and others to ask questions about the data. 

I'm thinking of using Open WebUI and oikb for the interface and data ingestions, and llama.ccp/vllm/ollama as engine (not sure yet), but I'd really like som advice on what current open weight models would be good for the ingestion and separately for the usage.

I'l be running it on mainly CPUs and if I can a few GPUs. What's best right now? Any recommendations?
```



document fetched 2026-09-28 00:39 · canonical https://reddit.com/r/LocalLLaMA/comments/1wrfjnc/advice_on_models_for_rag_use_case
