ragornot

RAG or not — decide.

Do you actually need RAG?

Eight quick questions, one clear recommendation — RAG, lexical search, long-context, fine-tuning, or nothing at all — each with the real cost and energy tradeoff. The logic is a fixed decision tree you can inspect, and it's happy to tell you that you don't need RAG. Curious how the two options compare live? Run the Benchmark.

0/8
1.What do you actually need out of this?

Be honest about the end goal — it's the biggest fork in the road.

2.Does answering require private, proprietary, or frequently-updated data the model wasn't trained on?
3.Do you need citations or verifiable grounding?

i.e. users must be able to trace an answer back to a source.

4.How much reference material is relevant to a typical question?
5.How often does that reference material change?
6.Is very low latency (sub-second) critical?
7.Is cost-per-query tightly constrained?

High volume, thin margins, or a hard budget per request.

8.Roughly how many queries per day?

Answer all 8questions to see a recommendation. There's no wrong answer — the tool is happy to tell you that you don't need RAG, or don't need AI at all.