Multimodal · Benchmark profile
A document understanding benchmark used in frontier-model comparison tables to measure extraction and grounded reasoning quality on complex documents.
Data verified 27 Jul 2026 · Methodology 1.8.0
Benchmark score on OmniDocBench 1.5
MiniMax M3 leads at 91.6%, followed by Qwen3.7-Plus (91.4%) and Qwen3.6-35B-A3B (89.9%).
3 ranked models · higher is better · labels show rank and score
| Model | Rank | Provider | Score |
|---|---|---|---|
| MiniMax M3 | #1 | MiniMax | 91.6% |
| Qwen3.7-Plus | #2 | Alibaba Cloud | 91.4% |
| Qwen3.6-35B-A3B | #3 | Alibaba Cloud | 89.9% |
One best score per model · higher is better
| Rank | Model | Provider | License | Evidence use | Score |
|---|---|---|---|---|---|
| #1 | MiniMax M3 MiniMax-M3 | MiniMax | open | Reference only | 91.6% |
| #2 | Qwen3.7-Plus qwen3.7-plus | Alibaba Cloud | closed | Reference only | 91.4% |
| #3 | Qwen3.6-35B-A3B qwen3-6-35b-a3b | Alibaba Cloud | open | Estimated reference | 89.9% |
The top of this snapshot is led by MiniMax M3 at 91.6%; third place is 1.7 points behind. The top-3 spread is 1.7 points.
About OmniDocBench 1.5
A document understanding benchmark used in frontier-model comparison tables to measure extraction and grounded reasoning quality on complex documents. Results stay tied to the exact model variant and evaluation system. Multiple systems for the same model use the best published score on this page; overall Lumina scoring uses the median of ranking-eligible rows.
Open benchmark source ↗FAQ
A document understanding benchmark used in frontier-model comparison tables to measure extraction and grounded reasoning quality on complex documents.
MiniMax M3 by MiniMax currently leads with 91.6%.
3 models in the LuminaBench cohort have a qualifying score on this benchmark.
No. This benchmark is display-only and does not enter the overall Lumina composite.
Related
