RIP Commercial #OCR. An #OpenSource #Model Just Topped Every Benchmark.
RIP Commercial OCR. An Open-Source Model Just Topped Every Benchmark.A 4B parameter model from a Brooklyn startup is beating GPT-4o and Gemini across 90 languages. Although OCR is not my domain, this one…
🔍 Hybrid unlocks: borderless table extraction (0.49−0.93 TEDS), #OCR in 80 languages, #LaTeX formula extraction from scientific papers, AI chart & image descriptions🛡️ Built-in #AI safety: fil…