Google Open-Sources Magika: AI File Type Detector at 99% Accuracy
alex_verem · x · 2026-08-02
Traditional file type detection relies heavily on extensions, which are easily spoofed. Google has open-sourced Magika, an AI-powered tool that identifies file types by reading their actual content.
- Technology: Employs a highly optimized deep learning model (just a few MBs) that runs in milliseconds on a single CPU.
- Performance: Trained on 100 million samples across 200+ file types, achieving 99% accuracy at 5ms per file.
- Scale: Already deployed internally for Gmail attachments and Google Drive, processing hundreds of billions of files weekly.
- Availability: Free and open-source on GitHub with 17.9k stars; runs entirely locally without cloud accounts.
More from coding & agent
- Memvid: Replace Complex RAG Pipelines with a Single-File Memory Layer for AI Agents — tom_doerr · 2026-08-02
- New RAG Project Uses Pre-Retrieval Cache to Cut Latency by 100x — sharpeye_wnl · 2026-08-02
- OpenAI Codex Caught Controlling Browser and Creating API Keys Without Permission — doodlestein · 2026-08-02
- Future AGI: Open-Source Platform for Evaluating and Shipping Self-Improving Agents — tom_doerr · 2026-08-02
- Shooting Short Films with Grok Imagine: 4 Skills for an Automated Workflow — tetsuoai · 2026-08-02
- MCP vs A2A vs Function Calling: Key Protocols for Production AI Agents — goyalshaliniuk · 2026-08-02