RAG & retrieval corpus
15 short, deterministic, non-overlapping articles make up this RAG testing dataset, built to exercise a chunking, embedding, and retrieval pipeline end to end. Each article is available as HTML, plain Markdown, or a generated PDF, and the whole set can be pulled in one call as JSON at /ai/corpus/articles.json. The full dataset, including every format's URL and the answer key, is also available as a single download at /ai/corpus/dataset.json. Looking for RAG testing PDFs built to stress-test extraction quality (tables, multi-column layout, repeated headers and footers)? See PDF documents.
Articles
- Photosynthesis · markdown · pdf
- The water cycle · markdown · pdf
- Firewalls · markdown · pdf
- Compound interest · markdown · pdf
- How vaccines work · markdown · pdf
- DNS · markdown · pdf
- The Pythagorean theorem · markdown · pdf
- How solar panels work · markdown · pdf
- Ocean tides · markdown · pdf
- HTTP vs HTTPS · markdown · pdf
- The four-stroke engine · markdown · pdf
- Inflation · markdown · pdf
- Noise-cancelling headphones · markdown · pdf
- The carbon cycle · markdown · pdf
- VPNs · markdown · pdf
Answer key
Run each query below through your retrieval pipeline. It should return the paired article as the top result.
| Query | Expected article |
|---|---|
| How do plants convert sunlight into energy? | Photosynthesis |
| Why does it rain? | The water cycle |
| What blocks unauthorized network traffic? | Firewalls |
| Why does my savings account grow faster over time? | Compound interest |
| How does the immune system learn to fight a disease before I catch it? | How vaccines work |
| How does my browser find a website's server? | DNS |
| How do I find the missing side of a right triangle? | The Pythagorean theorem |
| How does sunlight get turned into electricity? | How solar panels work |
| Why does the ocean rise and fall twice a day? | Ocean tides |
| Why do browsers say a site is not secure? | HTTP vs HTTPS |
| What are the four stages of a car engine's cycle? | The four-stroke engine |
| Why does money buy less over time? | Inflation |
| How do headphones block out engine noise? | Noise-cancelling headphones |
| How does carbon move between the air and living things? | The carbon cycle |
| How can I hide my IP address on public Wi-Fi? | VPNs |