# Text-to-Video Generation on VBench official evaluation prompts

Source: [Text-to-Video Generation on VBench official evaluation prompts](<https://www.sota2.com/research/sota/text-to-video-generation-on-vbench-official-evaluation-prompts>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-03-06T10:19:00.343Z

Evaluation of Text-to-Video Generation models using the VBench official evaluation prompts dataset, covering perceptual quality metrics (Semantic, Quality) and performance metrics (Latency, Speedup).

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 official evaluation prompts |
| Metric | Semantic Score |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 15 of 15 |
| 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

- [Semantic Score](<https://www.sota2.com/research/sota/text-to-video-generation-on-vbench-official-evaluation-prompts.md?metric=Semantic%20Score>): Higher is better; unit: SCALAR.

- [Quality Score](<https://www.sota2.com/research/sota/text-to-video-generation-on-vbench-official-evaluation-prompts.md?metric=Quality%20Score>): Higher is better; unit: SCALAR.

- [Total Score](<https://www.sota2.com/research/sota/text-to-video-generation-on-vbench-official-evaluation-prompts.md?metric=Total%20Score>): Higher is better; unit: SCALAR.

- [Sparsity](<https://www.sota2.com/research/sota/text-to-video-generation-on-vbench-official-evaluation-prompts.md?metric=Sparsity>): Lower is better; unit: PERCENT.

- [Latency (s)](<https://www.sota2.com/research/sota/text-to-video-generation-on-vbench-official-evaluation-prompts.md?metric=Latency%20(s)>): Lower is better; unit: TIME.

- [Speedup (Ratio)](<https://www.sota2.com/research/sota/text-to-video-generation-on-vbench-official-evaluation-prompts.md?metric=Speedup%20(Ratio)>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Semantic Score | Result date | Report paper |
| --- | --- | --- | --- |
| RadialAttention | 76.69 | Not recorded | 69aa5f9c2d603e6de3e41a9a |
| SpargeAttention | 72.87 | Not recorded | 69aa5f9c2d603e6de3e41a9a |
| CalibAtt | 72.81 | Not recorded | 69aa5f9c2d603e6de3e41a9a |
| CalibAtt | 72.8 | Not recorded | 69aa5f9c2d603e6de3e41a9a |
| Dense (FA3) | 72.71 | Not recorded | 69aa5f9c2d603e6de3e41a9a |
| RadialAttention | 72.44 | Not recorded | 69aa5f9c2d603e6de3e41a9a |
| Dense (FA3) | 71.65 | Not recorded | 69aa5f9c2d603e6de3e41a9a |
| SpargeAttention | 71.12 | Not recorded | 69aa5f9c2d603e6de3e41a9a |
| SVG2 | 70.03 | Not recorded | 69aa5f9c2d603e6de3e41a9a |
| SpargeAttention | 69.03 | Not recorded | 69aa5f9c2d603e6de3e41a9a |
| SVG2 | 67.84 | Not recorded | 69aa5f9c2d603e6de3e41a9a |
| CalibAtt | 67.34 | Not recorded | 69aa5f9c2d603e6de3e41a9a |
| Dense (FA3) | 67.02 | Not recorded | 69aa5f9c2d603e6de3e41a9a |
| RadialAttention | 66.16 | Not recorded | 69aa5f9c2d603e6de3e41a9a |
| SVG2 | 65.24 | Not recorded | 69aa5f9c2d603e6de3e41a9a |

### RadialAttention

Semantic Score: 76.69. Result ID: 69aaaa086dbb7075b954f108.

[Reporting paper](<https://www.sota2.com/research/paper/69aa5f9c2d603e6de3e41a9a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79ab1e209131cea7ecf7.md>)

Recorded evaluation configuration: {"Model":"Wan 2.1 14B","Resolution":"480p","Frames":"81 frames"}. This record does not assert independent reproduction.

### SpargeAttention

Semantic Score: 72.87. Result ID: 69aaaa086dbb7075b954f107.

[Reporting paper](<https://www.sota2.com/research/paper/69aa5f9c2d603e6de3e41a9a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bbe1e209131cea8a28a.md>)

Recorded evaluation configuration: {"Model":"Wan 2.1 14B","Resolution":"480p","Frames":"81 frames"}. This record does not assert independent reproduction.

### CalibAtt

Semantic Score: 72.81. Result ID: 69aaaa086dbb7075b954f10f.

[Reporting paper](<https://www.sota2.com/research/paper/69aa5f9c2d603e6de3e41a9a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69aa5f9c2d603e6de3e41a9a.md>)

Recorded evaluation configuration: {"Model":"Wan 2.1 14B","Resolution":"720p","Frames":"81 frames"}. This record does not assert independent reproduction.

### CalibAtt

Semantic Score: 72.8. Result ID: 69aaaa086dbb7075b954f10a.

[Reporting paper](<https://www.sota2.com/research/paper/69aa5f9c2d603e6de3e41a9a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69aa5f9c2d603e6de3e41a9a.md>)

Recorded evaluation configuration: {"Model":"Wan 2.1 14B","Resolution":"480p","Frames":"81 frames"}. This record does not assert independent reproduction.

### Dense (FA3)

Semantic Score: 72.71. Result ID: 69aaaa086dbb7075b954f106.

[Reporting paper](<https://www.sota2.com/research/paper/69aa5f9c2d603e6de3e41a9a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa740f1e209131cea5f89c.md>)

[Implementation](<https://github.com/Dao-AILab/flash-attention>)

Recorded evaluation configuration: {"Model":"Wan 2.1 14B","Resolution":"480p","Frames":"81 frames"}. This record does not assert independent reproduction.

### RadialAttention

Semantic Score: 72.44. Result ID: 69aaaa086dbb7075b954f10d.

[Reporting paper](<https://www.sota2.com/research/paper/69aa5f9c2d603e6de3e41a9a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79ab1e209131cea7ecf7.md>)

Recorded evaluation configuration: {"Model":"Wan 2.1 14B","Resolution":"720p","Frames":"81 frames"}. This record does not assert independent reproduction.

### Dense (FA3)

Semantic Score: 71.65. Result ID: 69aaaa086dbb7075b954f10b.

[Reporting paper](<https://www.sota2.com/research/paper/69aa5f9c2d603e6de3e41a9a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa740f1e209131cea5f89c.md>)

[Implementation](<https://github.com/Dao-AILab/flash-attention>)

Recorded evaluation configuration: {"Model":"Wan 2.1 14B","Resolution":"720p","Frames":"81 frames"}. This record does not assert independent reproduction.

### SpargeAttention

Semantic Score: 71.12. Result ID: 69aaaa086dbb7075b954f10c.

[Reporting paper](<https://www.sota2.com/research/paper/69aa5f9c2d603e6de3e41a9a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bbe1e209131cea8a28a.md>)

Recorded evaluation configuration: {"Model":"Wan 2.1 14B","Resolution":"720p","Frames":"81 frames"}. This record does not assert independent reproduction.

### SVG2

Semantic Score: 70.03. Result ID: 69aaaa086dbb7075b954f10e.

[Reporting paper](<https://www.sota2.com/research/paper/69aa5f9c2d603e6de3e41a9a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7be31e209131cea8b13e.md>)

Recorded evaluation configuration: {"Model":"Wan 2.1 14B","Resolution":"720p","Frames":"81 frames"}. This record does not assert independent reproduction.

### SpargeAttention

Semantic Score: 69.03. Result ID: 69aaaa086dbb7075b954f111.

[Reporting paper](<https://www.sota2.com/research/paper/69aa5f9c2d603e6de3e41a9a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bbe1e209131cea8a28a.md>)

Recorded evaluation configuration: {"Model":"Mochi 1","Resolution":"480p","Frames":"85 frames"}. This record does not assert independent reproduction.

### SVG2

Semantic Score: 67.84. Result ID: 69aaaa086dbb7075b954f109.

[Reporting paper](<https://www.sota2.com/research/paper/69aa5f9c2d603e6de3e41a9a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7be31e209131cea8b13e.md>)

Recorded evaluation configuration: {"Model":"Wan 2.1 14B","Resolution":"480p","Frames":"81 frames"}. This record does not assert independent reproduction.

### CalibAtt

Semantic Score: 67.34. Result ID: 69aaaa086dbb7075b954f114.

[Reporting paper](<https://www.sota2.com/research/paper/69aa5f9c2d603e6de3e41a9a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69aa5f9c2d603e6de3e41a9a.md>)

Recorded evaluation configuration: {"Model":"Mochi 1","Resolution":"480p","Frames":"85 frames"}. This record does not assert independent reproduction.

### Dense (FA3)

Semantic Score: 67.02. Result ID: 69aaaa086dbb7075b954f110.

[Reporting paper](<https://www.sota2.com/research/paper/69aa5f9c2d603e6de3e41a9a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa740f1e209131cea5f89c.md>)

[Implementation](<https://github.com/Dao-AILab/flash-attention>)

Recorded evaluation configuration: {"Model":"Mochi 1","Resolution":"480p","Frames":"85 frames"}. This record does not assert independent reproduction.

### RadialAttention

Semantic Score: 66.16. Result ID: 69aaaa086dbb7075b954f112.

[Reporting paper](<https://www.sota2.com/research/paper/69aa5f9c2d603e6de3e41a9a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79ab1e209131cea7ecf7.md>)

Recorded evaluation configuration: {"Model":"Mochi 1","Resolution":"480p","Frames":"85 frames"}. This record does not assert independent reproduction.

### SVG2

Semantic Score: 65.24. Result ID: 69aaaa086dbb7075b954f113.

[Reporting paper](<https://www.sota2.com/research/paper/69aa5f9c2d603e6de3e41a9a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7be31e209131cea8b13e.md>)

Recorded evaluation configuration: {"Model":"Mochi 1","Resolution":"480p","Frames":"85 frames"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.
