← Benchmarks
OmniDocBench 1.5
OmniDocBench 1.5 is a comprehensive benchmark for evaluating multimodal large language models on document understanding tasks, including OCR, document parsing, information extraction, and visual question answering across diverse document types. Lower Overall Edit Distance scores are better.
id omnidocbench-1.5 · max 1 · 20 models reported
| # | Model | Score |
|---|---|---|
| 1 | MiniMax M3 MiniMax · open | 0.92 |
| 2 | Qwen3.7-Plus Alibaba Cloud / Qwen Team | 0.91 |
| 3 | Qwen3.6 Plus Alibaba Cloud / Qwen Team | 0.91 |
| 4 | Qwen3.8-27B Alibaba Cloud / Qwen Team · open | 0.91 |
| 5 | Qwen3.6-35B-A3B Alibaba Cloud / Qwen Team · open | 0.90 |
| 6 | Qwen3.5-122B-A10B Alibaba Cloud / Qwen Team · open | 0.90 |
| 7 | Qwen3.5-35B-A3B Alibaba Cloud / Qwen Team · open | 0.89 |
| 8 | GPT-5.4 OpenAI | 0.89 |
| 9 | Qwen3.5-27B Alibaba Cloud / Qwen Team · open | 0.89 |
| 10 | Kimi K2.5 Moonshot AI · open | 0.89 |
| 11 | GPT-5.5 Instant OpenAI | 0.88 |
| 12 | GPT-5.4 mini OpenAI | 0.87 |
