multimodal benchmark
TVBench
TVBench is a temporal video understanding benchmark evaluating reasoning over actions, events, and temporal dynamics in videos.
Updated Aug 11, 2026
Models3
Model coverage3
MetricScore
EvidenceB
TVBench Ranking
Higher score ranks better on this benchmark.
| 01 | BY | 80.5% | 100.0% | 3 | C | |
| 02 | AC | 78.2% | 50.0% | 3 | C | |
| 03 | BY | 77.2% | 0.0% | 3 | C |
TVBench Score Distribution
A closer view of the leading scores on this benchmark.
TVBench Highlights
The leading models and scores on this benchmark.
What is TVBench?
What TVBench measures and how its scores work.
TVBench is a temporal video understanding benchmark evaluating reasoning over actions, events, and temporal dynamics in videos.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- TVBench
- Modality
- multimodal
- Primary category
- multimodal
- Score direction
- higher
- LLMBoard eligible
- Yes
- Evaluation key
- tvbench|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about TVBench.
Which model scores highest on TVBench?
Seed 2.1 Pro is currently ranked first with 80.5%.
What does TVBench measure?
TVBench is a temporal video understanding benchmark evaluating reasoning over actions, events, and temporal dynamics in videos.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
3 model results are currently shown.
Does this benchmark affect the overall score?
Yes. This benchmark can contribute to the current LLMBoard capability score.