Skip to main content
ModelScale

Multimodal & Grounded benchmark

OmniDocBench 1.6 leaderboard

OmniDocBench v1.6. Every model the catalog carries a published OmniDocBench 1.6 value for, ranked by that value.

CategoryMultimodal & Grounded
MeasureComposite of edit distance, CDM and TEDS
TasksDocument parsing and extraction
DifficultyGrounded document reasoning

A document parsing benchmark whose overall score composites edit distance, CDM and TEDS sub-scores across diverse PDF layouts.

OmniDocBench 1.6 ranking

1 model with a published OmniDocBench 1.6 value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.

Models ranked by their published OmniDocBench v1.6 value
RankModelProviderComposite of edit distance, CDM and TEDS
1PMTernary Bonsai 2 27BPrism ML89.1

Evidence key: Observed

Rows are ordered by the value OmniDocBench: Benchmarking Diverse PDF Document Parsing with Comprehensive Annotations published, highest first. The source does not state whether a higher value is the better result, so this page does not either: for a benchmark that measures a rate of failure, read the table from the bottom.

All leaderboards