# Short Video Generation on VBench-Long 30 seconds Source: [Short Video Generation on VBench-Long 30 seconds]() Collection index: [llms.txt]() Data updated: 2026-03-16T08:49:45.695Z Evaluation of models on the VBench-Long 30 seconds dataset, focusing on perceptual quality metrics including Aesthetic Quality, Subject/Background Consistency, Motion Smoothness, and Temporal Flickering. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Short Video Generation | | Dataset (including any stated split) | VBench-Long 30 seconds | | Metric | Aesthetic Quality | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 10 of 10 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Aesthetic Quality](): Higher is better; unit: PERCENT. - [Background Consistency](): Higher is better; unit: PERCENT. - [Imaging Quality](): Higher is better; unit: PERCENT. - [Motion Smoothness](): Higher is better; unit: PERCENT. - [Subject Consistency](): Higher is better; unit: PERCENT. - [Temporal Flickering](): Higher is better; unit: PERCENT. - [Average Score](): Higher is better; unit: PERCENT. - [Delta](): Higher is better; unit: SCALAR. ## Recorded results | Method | Aesthetic Quality | Result date | Report paper | | --- | --- | --- | --- | | Deep Forcing | 59.87 | Not recorded | 69b78e9d0636e5bf4afbdcc2 | | ∞-RoPE | 59.32 | Not recorded | 69b78e9d0636e5bf4afbdcc2 | | MemRoPE | 59.31 | Not recorded | 69b78e9d0636e5bf4afbdcc2 | | LongLive | 58.57 | Not recorded | 69b78e9d0636e5bf4afbdcc2 | | MemRoPE | 58.48 | Not recorded | 69b78e9d0636e5bf4afbdcc2 | | Rolling Forcing | 58.45 | Not recorded | 69b78e9d0636e5bf4afbdcc2 | | Self-Forcing | 58.28 | Not recorded | 69b78e9d0636e5bf4afbdcc2 | | Deep Forcing | 57.83 | Not recorded | 69b78e9d0636e5bf4afbdcc2 | | CausVid | 57.25 | Not recorded | 69b78e9d0636e5bf4afbdcc2 | | ∞-RoPE | 57.13 | Not recorded | 69b78e9d0636e5bf4afbdcc2 | ### Deep Forcing Aesthetic Quality: 59.87. Result ID: 69b7c41247f34ce9249f7ede. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"LongLive"}. This record does not assert independent reproduction. ### ∞-RoPE Aesthetic Quality: 59.32. Result ID: 69b7c41247f34ce9249f7edf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"LongLive"}. This record does not assert independent reproduction. ### MemRoPE Aesthetic Quality: 59.31. Result ID: 69b7c41247f34ce9249f7ee0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"LongLive"}. This record does not assert independent reproduction. ### LongLive Aesthetic Quality: 58.57. Result ID: 69b7c41247f34ce9249f7edd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"LongLive"}. This record does not assert independent reproduction. ### MemRoPE Aesthetic Quality: 58.48. Result ID: 69b7c41247f34ce9249f7edc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Self-Forcing"}. This record does not assert independent reproduction. ### Rolling Forcing Aesthetic Quality: 58.45. Result ID: 69b7c41247f34ce9249f7ed8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Self-Forcing Aesthetic Quality: 58.28. Result ID: 69b7c41247f34ce9249f7ed9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Self-Forcing"}. This record does not assert independent reproduction. ### Deep Forcing Aesthetic Quality: 57.83. Result ID: 69b7c41247f34ce9249f7eda. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Self-Forcing"}. This record does not assert independent reproduction. ### CausVid Aesthetic Quality: 57.25. Result ID: 69b7c41247f34ce9249f7ed7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### ∞-RoPE Aesthetic Quality: 57.13. Result ID: 69b7c41247f34ce9249f7edb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Self-Forcing"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.