Unlimited-OCR-2
Baidu's Unlimited-OCR-2 is an open-weights (MIT) OCR model for ultra-long document parsing using sparse reference sliding-window attention. Time is the Hugging Face repo creation timestamp.
Baidu ERNIEDocumentLaunched
Why it isn't ranked
A document-parsing OCR model; none of our ranked benchmarks apply.
Launches outside the ranked categories are tracked, not scored: they never appear on the charts or in the rankings.
Banner
On the “New in past 24 hours” banner for 24 hours after launch, until .
Official page
huggingface.co ↗Sources
“Ultra-Long Document Parsing SOTA: Achieves state-of-the-art performance across both dozens-of-pages benchmarks (e.g., MPDocBench) and extreme hundred-page benchmarks (e.g., MarathonDocBench).”