BenchmarkOmniDocBench v1.5
OmniDocBench v1.5
A benchmark for evaluating document parsing capabilities, covering a variety of document types and layouts.
Timeline (last 90 days)
Oct 13
Qwen3.8-27B scored 91.1 on OmniDocBench v1.5, the strongest document parsing score among the generative models in this evaluation.
Unverified50%