Agents
Messy tables and scanned PDFs are where most parsing breaks. OCRia-1 doesn't. Huge work from our research team, and it's already powering al…
Messy tables and scanned PDFs are where most parsing breaks. OCRia-1 doesn't. Huge work from our research team, and it's already powering all of LlamaParse🚀 Today we’re releasing OCRia-1, a frontier V
Messy tables and scanned PDFs are where most parsing breaks. OCRia-1 doesn't. Huge work from our research team, and it's already powering all of LlamaParse🚀 Today we’re releasing OCRia-1, a frontier VLM for document parsing that sets a new SOTA in accuracy, cost, and latency. It not only sets a new high score on ParseBench (87.4), but also leads in three 3rd-party benchmarks: olmOCR-bench* (83.5 overall, 92.9 on tables!), PulseBench-…
Related
- This is a huge moment for LlamaIndex and an incredible milestone for our research team: introducing OCRia-1, our new document parsing model.…
- llamaindex shipped OCRia-1, a little 4B model that reads ugly pdfs and turns tables, charts and formulas into clean data, beating parsers tw…
- LlamaParse is bringing their first kid into the world: OCRia-1 is here, and they really like to read (fast) The team went all in on this mod…
Source: Jerry Liu (X) | 2026-10-22