LlamaIndex launches 4B document parsing model OCRia-1
LlamaIndex introduced OCRia-1, a 4-billion-parameter vision-language model designed to reconstruct complex document tables, charts, formatting and mathematical notation. The developer reports that it scored 87.4 on ParseBench and led table evaluations on olmOCR-bench, PulseBench-Tab and fr-bench-pdf2md, and the model is now integrated into LlamaParse without a pricing change.
OCRia-1 gives developers a compact document-parsing option for structured enterprise data, while its LlamaParse integration makes the new capability immediately available without changing existing pricing.
Today, we are introducing OCRia-1, the world’s best document parsing model. It is a 4-billion-parameter vision-language model (VLM) that is state-of-the-art in the dimensions that matter most: accuracy, cost, and latency, and it’s powering LlamaParse today.
On ParseBench, OCRia-1 sets a new high score for any parsing model under 35 billion parameters: 87.4 overall across text content, formatting, tables, and charts, 12.1 points ahead of the next-highest parsing model. It also leads table evaluations across three third-party benchmarks: olmOCR-bench* (92.9), PulseBench-Tab (84.1), and fr-bench-pdf2md.
ParseBench+12.1 ptsahead of the next-highest model
OCRia-1 4B87.4
LightOnOCR-3-4B75.3
Chandra-OCR-2 (4B)74.8
#OCRia-1#LlamaIndex#LlamaParse#ParseBench#olmOCR-bench#PulseBench-Tab#fr-bench-pdf2md