multimodal benchmark
Video-MME Leaderboard
Video-MME is a multimodal benchmark for evaluating video analysis by MLLMs using 900 videos, 2,700 human-annotated question-answer pairs, and diverse video inputs.
Updated Sep 7, 2026
Models17
Model coverage17
MetricScore
EvidenceB
Video-MME Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelBY | Score89.20% | Percentile100.00% | Participants17 | EvidenceC | Evaluated |
| Rank02 | ModelBY | Score89.00% | Percentile93.75% | Participants17 | EvidenceC | Evaluated |
| Rank03 | ModelAC | Score88.00% | Percentile87.50% | Participants17 | EvidenceC | Evaluated |
| Rank04 | ModelXI | Score87.70% | Percentile81.25% | Participants17 | EvidenceC | Evaluated |
| Rank05 | ModelMA | Score87.40% | Percentile75.00% | Participants17 | EvidenceC | Evaluated |
| Rank06 | ModelMI | Score85.40% | Percentile68.75% | Participants17 | EvidenceC | Evaluated |
| Rank07 | ModelGO | Score84.80% | Percentile62.50% | Participants17 | EvidenceC | Evaluated |
| Rank08 | ModelAC | Score84.20% | Percentile56.25% | Participants17 | EvidenceC | Evaluated |
| Rank09 | ModelGO | Score78.60% | Percentile50.00% | Participants17 | EvidenceC | Evaluated |
| Rank10 | ModelAM Nova 2 OmniAmazon | Score77.90% | Percentile43.75% | Participants17 | EvidenceC | Evaluated |
| Rank11 | ModelGO | Score76.10% | Percentile37.50% | Participants17 | EvidenceC | Evaluated |
| Rank12 | ModelAC | Score74.50% | Percentile31.25% | Participants17 | EvidenceC | Evaluated |
| Rank13 | ModelAC | Score73.30% | Percentile25.00% | Participants17 | EvidenceC | Evaluated |
| Rank14 | ModelAC | Score71.80% | Percentile18.75% | Participants17 | EvidenceC | Evaluated |
| Rank15 | ModelAC | Score71.40% | Percentile12.50% | Participants17 | EvidenceC | Evaluated |
| Rank16 | ModelGO | Score66.20% | Percentile6.25% | Participants17 | EvidenceC | Evaluated |
| Rank17 | ModelMI | Score55.00% | Percentile0.00% | Participants17 | EvidenceC | Evaluated |
Video-MME Highlights
The leading models and scores on this benchmark.
Video-MME Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for Video-MME
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What is Video-MME?
What Video-MME measures and how its scores work.
Video-MME is a benchmark for evaluating Multi-modal Large Language Models (MLLMs) in video analysis, covering 6 primary visual domains and 30 subfields.
It measures model performance on video-analysis questions across temporal dimensions from 11 seconds to 1 hour using video frames, subtitles, and audio, with the metric reported as Score in ratio units.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- Video-MME
- 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 Video-MME.
Which model scores highest on Video-MME?
Seed 2.1 Pro is currently ranked first with 89.20%.
What are the top three models on Video-MME?
The current leaders are Seed 2.1 Pro (89.20%), Seed 2.1 Turbo (89.00%), and Qwen3.7-Plus (88.00%).
Which Video-MME model has the lowest official input price?
MiMo-V2.5 has the lowest matched official input price at $0.14 input / $0.28 output per 1M tokens.
Which models are fastest among Video-MME results?
The fastest matched records are Gemini 1.5 Flash (150.00 tok/s via Google), Gemini 1.5 Flash 8B (150.00 tok/s via Google), and Gemini 1.5 Pro (85.00 tok/s via Google).
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.