NewPage Search API — cut 97% token cost. Learn more
Get verifiable answers from complex documents
Get faster, more accurate answers at lower cost, backed by evidence from your documents. No embeddings. No vector database.
Explore PageIndex AgentPage Search API
Page Search API finds the relevant pages for each question in about a second, so your LLM reads only those, not the whole PDF.
- 97%
- less token cost
- ~1s
- Including index & search, faster and more accurate than SOTA vector search
Page Extract API
Extract the data out of any PDF: describe the fields you want in a JSON Schema, and get them back from the document.
- 60%
- lower cost than the next best
- #1
- in accuracy on ExtractBench
{
"holdings": [
{
"issuer": "111 INC",
"value": 84851
},
… 3,062 more
]
}