# Text-to-Video on 1,024 prompts (held-out)

Source: [Text-to-Video on 1,024 prompts (held-out)](<https://www.sota2.com/research/sota/text-to-video-on-1-024-prompts-held-out>)

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

Data updated: 2026-02-26T06:08:03.374Z

Evaluation of Text-to-Video models using the held-out set of 1,024 prompts, assessed using VQ and MQ metrics.

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 |
| Dataset (including any stated split) | 1,024 prompts (held-out) |
| Metric | VQ |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 5 of 5 |
| 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

- [VQ](<https://www.sota2.com/research/sota/text-to-video-on-1-024-prompts-held-out.md?metric=VQ>): Higher is better; unit: SCALAR.

- [MQ](<https://www.sota2.com/research/sota/text-to-video-on-1-024-prompts-held-out.md?metric=MQ>): Higher is better; unit: SCALAR.

## Recorded results

| Method | VQ | Result date | Report paper |
| --- | --- | --- | --- |
| MapReduce LoRA | 4.81 | Not recorded | 692689c5d6b0dd3342930642 |
| DanceGRPO | 4.37 | Not recorded | 692689c5d6b0dd3342930642 |
| Rewarded Soup | 4.13 | Not recorded | 692689c5d6b0dd3342930642 |
| DanceGRPO | 3.8 | Not recorded | 692689c5d6b0dd3342930642 |
| HunyuanVideo | 3.25 | Not recorded | 692689c5d6b0dd3342930642 |

### MapReduce LoRA

VQ: 4.81. Result ID: 699a3c8b5698665ec4c30e83.

[Reporting paper](<https://www.sota2.com/research/paper/692689c5d6b0dd3342930642.md>)

[Method origin paper](<https://www.sota2.com/research/paper/692689c5d6b0dd3342930642.md>)

Recorded evaluation configuration: {"Reward":"VQ+MQ"}. This record does not assert independent reproduction.

### DanceGRPO

VQ: 4.37. Result ID: 699a3c8b5698665ec4c30e80.

[Reporting paper](<https://www.sota2.com/research/paper/692689c5d6b0dd3342930642.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7afd1e209131cea8618d.md>)

Recorded evaluation configuration: {"Reward":"VQ"}. This record does not assert independent reproduction.

### Rewarded Soup

VQ: 4.13. Result ID: 699a3c8b5698665ec4c30e82.

[Reporting paper](<https://www.sota2.com/research/paper/692689c5d6b0dd3342930642.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70e61e209131cea4dcbb.md>)

Recorded evaluation configuration: {"Reward":"VQ+MQ"}. This record does not assert independent reproduction.

### DanceGRPO

VQ: 3.8. Result ID: 699a3c8b5698665ec4c30e81.

[Reporting paper](<https://www.sota2.com/research/paper/692689c5d6b0dd3342930642.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7afd1e209131cea8618d.md>)

Recorded evaluation configuration: {"Reward":"MQ"}. This record does not assert independent reproduction.

### HunyuanVideo

VQ: 3.25. Result ID: 699a3c8b5698665ec4c30e7f.

[Reporting paper](<https://www.sota2.com/research/paper/692689c5d6b0dd3342930642.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77581e209131cea72919.md>)

Recorded evaluation configuration: {"Reward":"None"}. This record does not assert independent reproduction.

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