Apify builder
Small Hours Small data tools for AI pipelines: PDFs, audio and recall data in, clean JSON out.
Leaderboard position
out of 10,354
New actors
#1,443
Total actors
#3,292
↓ -95
Monthly users
#3,696
↓ -80
Total users
#5,238
↓ -112
Monthly runs
#5,005
↑ +3,305
Total runs
#5,010
↑ +2,421
Portfolio stats
New actors
4
Total actors
4
↑ +1
Monthly users
4
↑ +1
Total users
8
↑ +2
Monthly runs
55
↑ +54
Total runs
278
↑ +234
New actors
Each bar shows one day in the last 30 days.
Portfolio concentration
4 of 4 actors cover 80%+ of monthly users.
Use to transcribe an audio file to text with timestamps, convert a video to subtitles in SRT or VTT, or turn a podcast episode into a transcript for RAG. Input: audioUrls (direct file links). Open-source Whisper. $0.012 per minute ($0.035 accurate). Not for YouTube links.
25%
users
Use to extract tables from a PDF as JSON (rows, page, cell coordinates), convert a PDF to text chunks for a RAG pipeline with page citations, or OCR a scanned PDF and return the text. Input: pdfUrls. $0.003 per page, +$0.01 per OCR page. Not for Word/Excel files or audio.
25%
users
Use to check if a product has been recalled by the USDA, FDA, CPSC or NHTSA, find recalled products by UPC or brand name, or get safety recalls for a car by make, model and year. One JSON schema. Input: sinceDays, query, upc, firm. $0.005 per recall record. Not for EU/UK recalls.
25%
users
Use to convert a Word document to text for an LLM, extract text from Excel and PowerPoint files, or read text from an image with OCR (also CSV and HTML), with RAG chunks and citations. Input: fileUrls. $0.003 per unit, +$0.01 per OCR image. Not for PDFs or old .doc/.xls files.
25%
users
ActorHunt MCP
ActorHunt gives paid members MCP access to keyword data, actor evidence, and Store movement for deciding what to build next.
