Paper Search Semantic search across paper knowledge cards. Vectors are built from title + abstract + summary + tags. Returns card-level metadata (not full text) so you can decide which papers to read in full.
Cost: 1 credit per callQuery limit: 2,000 characters max
Endpoint POST https://api.llmquantdata.com/api/paper/search
Authentication Include a Bearer token in the Authorization header.
Authorization: Bearer YOUR_API_KEY
Request Body Search query string (e.g. momentum crash, factor zoo, deep learning alpha). Max 2,000 characters.
Maximum number of results to return. Default 5, max 10.
Response The response follows the standard {data, meta} envelope.
Array of paper_search_result objects ranked by relevance. Unique identifier for this paper card. Use this with Paper Read to retrieve sections.
Identifier from the original source (e.g. arXiv ID).
Original abstract of the paper.
Curated summary of the paper’s key findings.
Sections available for reading via Paper Read. Key used to request this section in Paper Read.
Display title of the section.
Character count of the section content.
Order of the section within the paper.
Total number of available sections.
Total character count of the full paper text.
Direct URL to the paper PDF.
The topK value used for this request.
Credits consumed by this call. Always 1 for paper search.
Credits remaining on your account after this call.
Code Examples import requests url = "https://api.llmquantdata.com/api/paper/search" headers = { "Authorization" : "Bearer YOUR_API_KEY" , "Content-Type" : "application/json" , } response = requests.post(url, headers = headers, json = { "query" : "momentum crash" , "topK" : 5 , }) result = response.json() for paper in result[ "data" ]: print (paper[ "title" ]) print ( f " Authors: { ', ' .join(paper[ 'authors' ]) } " ) print ( f " Sections: { paper[ 'section_count' ] } , Chars: { paper[ 'full_text_char_count' ] } " ) print ( f " Tags: { paper[ 'tags' ] } " ) print () print ( f "Credits remaining: { result[ 'meta' ][ 'remainingCredits' ] } " )
Paper Read Read specific sections (or the full text) of a paper by its paper_card_id. Use this after search to retrieve the actual content.
Endpoint POST https://api.llmquantdata.com/api/paper/read
Authentication Include a Bearer token in the Authorization header.
Authorization: Bearer YOUR_API_KEY
Request Body The paper_card_id from a Paper Search result.
List of section_key values to read. Omit or pass ["all"] to read the full paper.
Response Unique identifier for this paper card.
Identifier from the original source.
Original abstract of the paper.
Curated summary of the paper’s key findings.
Direct URL to the paper PDF.
All sections available in this paper (same structure as search results).
Total number of available sections.
Total character count of the full paper text.
The requested sections with their Markdown content. Key identifying this section.
Display title of the section.
Section content in Markdown format.
Character count of the section content.
Order of the section within the paper.
Credits remaining on your account.
Code Examples Read Specific Sections
Read Full Paper
Node.js
cURL
import requests url = "https://api.llmquantdata.com/api/paper/read" headers = { "Authorization" : "Bearer YOUR_API_KEY" , "Content-Type" : "application/json" , } # Read only the introduction and methodology sections response = requests.post(url, headers = headers, json = { "paperCardId" : "card_abc123" , "sections" : [ "introduction" , "methodology" ], }) result = response.json() for section in result[ "data" ][ "sections" ]: print ( f "## { section[ 'title' ] } ( { section[ 'char_count' ] } chars)" ) print (section[ "content" ]) print ()