# Text-to-Video Generation on VBench T2V 15 Source: [Text-to-Video Generation on VBench T2V 15]() Collection index: [llms.txt]() Data updated: 2026-04-21T07:34:43.183Z Evaluation of Text-to-Video Generation models using the VBench T2V 15 dataset/configuration. This benchmark uses a comprehensive set of 19 metrics covering overall performance, perceptual quality, semantic alignment, temporal dynamics, and detailed consistency checks across various visual attributes. 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 | Text-to-Video Generation | | Dataset (including any stated split) | VBench T2V 15 | | Metric | Total Score | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 17 of 17 | | 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 - [Total Score](): Higher is better; unit: PERCENT. - [Quality Score](): Higher is better; unit: PERCENT. - [Semantic Score](): Higher is better; unit: PERCENT. - [Subject Consistency](): Higher is better; unit: PERCENT. - [Background Consistency](): Higher is better; unit: PERCENT. - [Temporal Flickering Score](): Higher is better; unit: PERCENT. - [Motion Smoothness](): Higher is better; unit: PERCENT. - [Dynamic Degree](): Higher is better; unit: PERCENT. - [Aesthetic Quality](): Higher is better; unit: PERCENT. - [Imaging Quality](): Higher is better; unit: PERCENT. - [Object Class Accuracy](): Higher is better; unit: PERCENT. - [Multiple Objects Accuracy](): Higher is better; unit: PERCENT. - [Human Action Accuracy](): Higher is better; unit: PERCENT. - [Color Fidelity Score](): Higher is better; unit: PERCENT. - [Spatial Relationship Accuracy](): Higher is better; unit: PERCENT. - [Scene Consistency Score](): Higher is better; unit: PERCENT. - [Appearance Style Fidelity](): Higher is better; unit: PERCENT. - [Temporal Style Fidelity](): Higher is better; unit: PERCENT. - [Overall Consistency Score](): Higher is better; unit: PERCENT. ## Recorded results | Method | Total Score | Result date | Report paper | | --- | --- | --- | --- | | Veo 3 | 85.06 | Not recorded | 69e7048e7d0172140ec7ecd0 | | Sora | 84.28 | Not recorded | 69e7048e7d0172140ec7ecd0 | | Wan2.2-T2V | 84.23 | Not recorded | 69e7048e7d0172140ec7ecd0 | | Motif-Video 2B | 83.76 | Not recorded | 69e7048e7d0172140ec7ecd0 | | SANA-Video | 83.71 | Not recorded | 69e7048e7d0172140ec7ecd0 | | Wan2.1-T2V | 83.69 | Not recorded | 69e7048e7d0172140ec7ecd0 | | Luma | 83.61 | Not recorded | 69e7048e7d0172140ec7ecd0 | | OpenSora 2.0 | 83.6 | Not recorded | 69e7048e7d0172140ec7ecd0 | | HunyuanVideo | 83.43 | Not recorded | 69e7048e7d0172140ec7ecd0 | | MiniMax-Video-01 | 83.41 | Not recorded | 69e7048e7d0172140ec7ecd0 | | Kling 1.6 | 83.4 | Not recorded | 69e7048e7d0172140ec7ecd0 | | Wan2.1-T2V | 83.31 | Not recorded | 69e7048e7d0172140ec7ecd0 | | HunyuanVideo | 83.24 | Not recorded | 69e7048e7d0172140ec7ecd0 | | CogVideoX1.5-5B | 82.17 | Not recorded | 69e7048e7d0172140ec7ecd0 | | CogVideoX-5B | 81.91 | Not recorded | 69e7048e7d0172140ec7ecd0 | | Step-Video-T2V | 81.83 | Not recorded | 69e7048e7d0172140ec7ecd0 | | LTX-Video | 80 | Not recorded | 69e7048e7d0172140ec7ecd0 | ### Veo 3 Total Score: 85.06. Result ID: 69e7281947f34ce924cd039a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Access":"Closed-source"}. This record does not assert independent reproduction. ### Sora Total Score: 84.28. Result ID: 69e7281947f34ce924cd039b. [Reporting paper]() Recorded evaluation configuration: {"Access":"Closed-source"}. This record does not assert independent reproduction. ### Wan2.2-T2V Total Score: 84.23. Result ID: 69e7281947f34ce924cd03a0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Params":"A14B","Prompt optimisation":"Qwen-rewritten prompts"}. This record does not assert independent reproduction. ### Motif-Video 2B Total Score: 83.76. Result ID: 69e7281947f34ce924cd03aa. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Params":"2B"}. This record does not assert independent reproduction. ### SANA-Video Total Score: 83.71. Result ID: 69e7281947f34ce924cd03a1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Params":"2B"}. This record does not assert independent reproduction. ### Wan2.1-T2V Total Score: 83.69. Result ID: 69e7281947f34ce924cd03a2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Params":"14B"}. This record does not assert independent reproduction. ### Luma Total Score: 83.61. Result ID: 69e7281947f34ce924cd039c. [Reporting paper]() Recorded evaluation configuration: {"Access":"Closed-source"}. This record does not assert independent reproduction. ### OpenSora 2.0 Total Score: 83.6. Result ID: 69e7281947f34ce924cd03a3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Params":"11B","Pipeline":"T2I2V (FLUX anchor frame)"}. This record does not assert independent reproduction. ### HunyuanVideo Total Score: 83.43. Result ID: 69e7281947f34ce924cd039d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Access":"Closed-source","Version":"Updated API checkpoint (2025-05-22)"}. This record does not assert independent reproduction. ### MiniMax-Video-01 Total Score: 83.41. Result ID: 69e7281947f34ce924cd039e. [Reporting paper]() Recorded evaluation configuration: {"Access":"Closed-source"}. This record does not assert independent reproduction. ### Kling 1.6 Total Score: 83.4. Result ID: 69e7281947f34ce924cd039f. [Reporting paper]() Recorded evaluation configuration: {"Access":"Closed-source"}. This record does not assert independent reproduction. ### Wan2.1-T2V Total Score: 83.31. Result ID: 69e7281947f34ce924cd03a4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Params":"1.3B"}. This record does not assert independent reproduction. ### HunyuanVideo Total Score: 83.24. Result ID: 69e7281947f34ce924cd03a5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Params":"13B"}. This record does not assert independent reproduction. ### CogVideoX1.5-5B Total Score: 82.17. Result ID: 69e7281947f34ce924cd03a6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Params":"5B","Prompt optimisation":"SAT-enhanced"}. This record does not assert independent reproduction. ### CogVideoX-5B Total Score: 81.91. Result ID: 69e7281947f34ce924cd03a7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Params":"5B"}. This record does not assert independent reproduction. ### Step-Video-T2V Total Score: 81.83. Result ID: 69e7281947f34ce924cd03a8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Params":"30B"}. This record does not assert independent reproduction. ### LTX-Video Total Score: 80. Result ID: 69e7281947f34ce924cd03a9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Params":"2B"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.