VEFX-Leaderboard
The standard benchmark for evaluating video editing models. Submit your results on 300 videos across 9 categories and get scored on three quality dimensions.
Three Evaluation Dimensions
Every submission is evaluated along three complementary dimensions, each scored on a 1–4 scale.
Instructional Following (IF)
Measures how well the edited video follows the editing instruction.
Range: 1-4 · Higher is better
Render Quality (RQ)
Measures the visual rendering quality of the edited video.
Range: 1-4 · Higher is better
Edit Exclusivity (EE)
Measures whether only the intended region/attribute was edited without side effects.
Range: 1-4 · Higher is better
How It Works
Four simple steps from benchmark download to detailed evaluation results.
Download Benchmark
Get the VEFX-Bench dataset with 300 source videos and editing instructions from HuggingFace.
Edit Videos
Apply your video editing model to produce 300 edited videos following the provided instructions.
Submit Results
Upload your edited videos as a .zip file. Our system validates and queues them for evaluation.
View Results
See detailed scores across IF, RQ, and EE dimensions with per-video breakdowns and category analysis.
Ready to evaluate your model?
Download the benchmark dataset, run your model, and submit your results to see how you compare.