multimodal benchmark
Video-MME (long, no subtitles)
Video-MME is the first-ever comprehensive evaluation benchmark for Multi-modal Large Language Models (MLLMs) in video analysis. This variant focuses on long-term videos (30min-60min) without subtitle inputs, testing robust contextual dynamics across 6 primary visual domains with 30 subfields including knowledge, film & television, sports competition, life record, and multilingual content.
Updated Aug 7, 2026
Published models1
Registry coverage1
MetricScore
EvidenceB
Video-MME (long, no subtitles) leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | OP | 72.0% | 100.0% | 1 | C |
Video-MME (long, no subtitles) highlights
The top published results on this benchmark's own scale.
What is Video-MME (long, no subtitles)?
Definition and scoring fields from the benchmark registry.
Video-MME is the first-ever comprehensive evaluation benchmark for Multi-modal Large Language Models (MLLMs) in video analysis. This variant focuses on long-term videos (30min-60min) without subtitle inputs, testing robust contextual dynamics across 6 primary visual domains with 30 subfields including knowledge, film & television, sports competition, life record, and multilingual content.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- Video-MME (long, no subtitles)
- Modality
- multimodal
- Primary category
- multimodal
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- video-mme-(long,-no-subtitles)|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about Video-MME (long, no subtitles).
Which model scores highest on Video-MME (long, no subtitles)?
GPT-4.1 is currently ranked first with 72.0%.
What does Video-MME (long, no subtitles) measure?
Video-MME is the first-ever comprehensive evaluation benchmark for Multi-modal Large Language Models (MLLMs) in video analysis. This variant focuses on long-term videos (30min-60min) without subtitle inputs, testing robust contextual dynamics across 6 primary visual domains with 30 subfields including knowledge, film & television, sports competition, life record, and multilingual content.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
1 unique published model results are currently shown.
Does this benchmark affect the overall score?
This benchmark is preserved as source-native evidence but is not eligible for the current overall score.