Skip to main content

MMLongBench-Doc

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

How BenchLM shows MMLongBench-Doc right now

BenchLM is tracking MMLongBench-Doc in the local dataset, but exact-source verification records for these rows are still being attached. To avoid a blank benchmark page, BenchLM shows the current tracked rows below as a display-only reference table.

These tracked rows are useful for inspection and spot-checking, but until exact-source attachments are completed they should not be treated as fully verified public benchmark rows.

4 tracked modelsLocal tracked rowsAwaiting exact-source attachmentsDisplay only

Tracked score on MMLongBench-Doc — April 10, 2026

BenchLM mirrors the published tracked score view for MMLongBench-Doc. Qwen3.6 Plus leads the public snapshot at 62.0% , followed by Qwen3.5 397B (61.5%) and Gemini 3 Pro (60.5%). BenchLM does not use these results to rank models overall.

4 modelsMultimodal & GroundedCurrentDisplay onlyUpdated April 10, 2026

The published MMLongBench-Doc snapshot is tightly clustered at the top: Qwen3.6 Plus sits at 62.0%, while the third row is only 1.5 points behind. The broader top-10 spread is 3.5 points, so many of the published scores sit in a relatively narrow band.

4 models have been evaluated on MMLongBench-Doc. The benchmark falls in the Multimodal & Grounded category. This category carries a 12% weight in BenchLM.ai's overall scoring system. MMLongBench-Doc is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.

About MMLongBench-Doc

Year

2026

Tasks

Long document understanding

Format

Document-grounded reasoning

Difficulty

Long-context document reasoning

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

BenchLM freshness & provenance

Version

MMLongBench-Doc 2026

Refresh cadence

Quarterly

Staleness state

Current

Question availability

Public benchmark set

CurrentDisplay only

BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.

Tracked score table (4 models)

1
Qwen3.6 Plusqwen3-6-plus
62.0%
2
Qwen3.5 397Bqwen3-5-397b
61.5%
3
Gemini 3 Progemini-3-pro
60.5%
4
Kimi K2.5kimi-k2-5
58.5%

FAQ

What does MMLongBench-Doc measure?

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

Which model leads the published MMLongBench-Doc snapshot?

Qwen3.6 Plus currently leads the published MMLongBench-Doc snapshot with a tracked score of 62.0%. BenchLM shows this benchmark for display only and does not use it in overall rankings.

How many models are evaluated on MMLongBench-Doc?

4 AI models are included in BenchLM's mirrored MMLongBench-Doc snapshot, based on the public leaderboard captured on April 10, 2026.

Last updated: April 10, 2026 · mirrored from the public benchmark leaderboard

The AI models change fast. We track them for you.

For engineers, researchers, and the plain curious — a weekly brief on new models, ranking shifts, and pricing changes.

Free. No spam. Unsubscribe anytime.