language benchmark
VATEX Leaderboard
VATEX is a multilingual video-and-language dataset containing over 41,250 videos, 825,000 captions in English and Chinese, and over 206,000 English-Chinese parallel translation pairs.
Updated Sep 5, 2026
Models2
Model coverage2
MetricScore
EvidenceB
VATEX Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelAM Nova LiteAmazon | Score77.80% | Percentile100.00% | Participants2 | EvidenceC | Evaluated |
| Rank02 | ModelAMNova ProAmazon | Score77.80% | Percentile0.00% | Participants2 | EvidenceC | Evaluated |
VATEX Highlights
The leading models and scores on this benchmark.
VATEX Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for VATEX
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What is VATEX?
What VATEX measures and how its scores work.
VATEX is a large-scale dataset for video-and-language research that supports multilingual video captioning and video-guided machine translation tasks.
VATEX reports a Score metric as a ratio.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- VATEX
- Modality
- multimodal
- Primary category
- language
- 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 VATEX.
Which model scores highest on VATEX?
Nova Lite is currently ranked first with 77.80%.
What are the top three models on VATEX?
The current leaders are Nova Lite (77.80%) and Nova Pro (77.80%).
Which VATEX model has the lowest official input price?
Nova Lite has the lowest matched official input price at $0.06 input / $0.24 output per 1M tokens.
Which models are fastest among VATEX 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.