Multimodal & Grounded benchmark
MMLongBench-Doc leaderboard
Every model the catalog carries a published MMLongBench-Doc value for, ranked by that value.
CategoryMultimodal & Grounded
MeasureDocument-grounded reasoning
TasksLong document understanding
DifficultyLong-context document reasoning
Published byQwen3.6 launch benchmarks
A long-document multimodal benchmark for grounded reasoning over extended document contexts.
MMLongBench-Doc ranking
1 model with a published MMLongBench-Doc value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.
| Rank | Model | Provider | Document-grounded reasoning |
|---|---|---|---|
| 1 |  Nemotron 3 Nano Omni 30B A3B | NVIDIA | 57.5 |
Evidence key: Observed
Rows are ordered by the value Qwen3.6 launch benchmarks 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.