multimodal benchmark
MMT-Bench Leaderboard
MMT-Bench is a multimodal benchmark with 31,325 multi-choice visual questions covering 32 core meta-tasks and 162 subtasks across scenarios including vehicle driving and embodied navigation.
Updated Sep 5, 2026
Models4
Model coverage4
MetricScore
EvidenceB
MMT-Bench Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelDE | Score63.60% | Percentile100.00% | Participants4 | EvidenceC | Evaluated |
| Rank02 | ModelAC | Score63.60% | Percentile66.67% | Participants4 | EvidenceC | Evaluated |
| Rank03 | ModelDE | Score62.90% | Percentile33.33% | Participants4 | EvidenceC | Evaluated |
| Rank04 | ModelDE | Score53.20% | Percentile0.00% | Participants4 | EvidenceC | Evaluated |
MMT-Bench Highlights
The leading models and scores on this benchmark.
MMT-Bench Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for MMT-Bench
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What is MMT-Bench?
What MMT-Bench measures and how its scores work.
MMT-Bench is a multimodal benchmark for evaluating Large Vision-Language Models on multimodal understanding.
It measures multimodal understanding using Score, reported as a ratio, across 32 core meta-tasks and 162 subtasks.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- MMT-Bench
- Modality
- multimodal
- Primary category
- multimodal
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- overall
LLM Stats. Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about MMT-Bench.
Which model scores highest on MMT-Bench?
DeepSeek VL2 is currently ranked first with 63.60%.
What are the top three models on MMT-Bench?
The current leaders are DeepSeek VL2 (63.60%), Qwen2.5 VL 7B Instruct (63.60%), and DeepSeek VL2 Small (62.90%).
Which MMT-Bench model has the lowest official input price?
Qwen2.5 VL 7B Instruct has the lowest matched official input price at $0.35 input / $1.1 output per 1M tokens.
Which models are fastest among MMT-Bench results?
No matched runtime record is currently available.
Does the highest score result prove overall model quality?
No. This benchmark measures one defined capability or task. The overall LLMBoard score uses a separate aggregation across eligible benchmark evidence.