Open Sourced Rust Tools Parse 14 Document Formats at ~5ms/Page
devdigest · x · 2026-08-12
The developer recently open-sourced two document processing tools built in Rust:
- anydoc: Parses 14 document formats at approximately 5ms per page.
- pdf-inspector: Parses and classifies PDFs without waiting for OCR.
Both tools have reached 14k stars on GitHub and now power Firecrawl's /parse endpoint, adding OCR capabilities for more complex scenarios.
More from coding & agent
- Omnigent Lets You Run AI Agents and Approve Tool Calls Directly in Slack — matei_zaharia · 2026-08-12
- Agent Loop Update: Verifying Work, Token Funding, and Revenue Routing — seanwbren · 2026-08-12
- MCP Usage Surges: Linear Sees Agent-Created Issues Surpass App-Created Ones — jeff_weinstein · 2026-08-12
- FlashRT: AI Agents Auto-Optimize Multimodal Deployment, Cutting Latency by 70x — BeidiChen · 2026-08-12
- xAI Launches Grok Bot Beta: AI Teammates That Operate Tools — xiaosun86 · 2026-08-12
- AI Agent on Break: Learns Spacecraft Tricks, Codes Simulation for Fun — max_paperclips · 2026-08-12