# MMLongBench-Doc

> A long-document multimodal benchmark for grounded reasoning over extended document contexts.

Canonical page: https://benchlm.ai/benchmarks/mmlongbenchdoc

- Category: [Multimodal & Grounded](/multimodal-grounded)
- Last updated: September 27, 2026

## About MMLongBench-Doc

- Year: 2026
- Tasks: Long document understanding
- Format: Document-grounded reasoning
- Difficulty: Long-context document reasoning
- Paper: [Qwen3.6 launch benchmarks](https://qwen.ai/blog?id=qwen3.6)

MMLongBench-Doc is designed to test whether a model can maintain grounded understanding across large document contexts rather than only short OCR-style snippets.

MMLongBench-Doc is currently displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.

## Leaderboard (1 models)

| Rank | Model | Creator | Score |
|------|-------|---------|-------|
| 1 | [Nemotron 3 Nano Omni 30B A3B](/models/nemotron-3-nano-omni-30b-a3b) | NVIDIA | 57.5% |

## FAQ

### What does MMLongBench-Doc measure?

A long-document multimodal benchmark for grounded reasoning over extended document contexts.

### Which model scores highest on MMLongBench-Doc?

Nemotron 3 Nano Omni 30B A3B by NVIDIA currently leads with a score of 57.5% on MMLongBench-Doc.

### How many models are evaluated on MMLongBench-Doc?

1 AI models have been evaluated on MMLongBench-Doc on BenchLM.

### Does MMLongBench-Doc affect BenchLM's overall score?

Not directly. MMLongBench-Doc is still displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.