# Video Generation on Panda-70M 1024 samples

Source: [Video Generation on Panda-70M 1024 samples](<https://www.sota2.com/research/sota/video-generation-on-panda-70m-1024-samples>)

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

Data updated: 2026-02-26T06:04:50.098Z

Evaluation of video generation models on the Panda-70M dataset (1024 samples subset), assessed using Structural Consistency (SC), Boundary Consistency (BC), Motion Smoothness (MS), and Fréchet Video Distance (FVD).

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 | Video Generation |
| Dataset (including any stated split) | Panda-70M 1024 samples |
| Metric | Structural Consistency |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 6 of 6 |
| 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

- [Structural Consistency](<https://www.sota2.com/research/sota/video-generation-on-panda-70m-1024-samples.md?metric=Structural%20Consistency>): Higher is better; unit: PERCENT.

- [Boundary Consistency](<https://www.sota2.com/research/sota/video-generation-on-panda-70m-1024-samples.md?metric=Boundary%20Consistency>): Higher is better; unit: PERCENT.

- [Motion Smoothness](<https://www.sota2.com/research/sota/video-generation-on-panda-70m-1024-samples.md?metric=Motion%20Smoothness>): Higher is better; unit: PERCENT.

- [Fréchet Video Distance (FVD)](<https://www.sota2.com/research/sota/video-generation-on-panda-70m-1024-samples.md?metric=Fr%C3%A9chet%20Video%20Distance%20(FVD)>): Lower is better; unit: SCALAR.

## Recorded results

| Method | Structural Consistency | Result date | Report paper |
| --- | --- | --- | --- |
| UDPDiff | 97.07 | Not recorded | 68fa77681e209131cea73147 |
| UDPDiff | 95.91 | Not recorded | 68fa77681e209131cea73147 |
| UDPDiff | 95.21 | Not recorded | 68fa77681e209131cea73147 |
| UDPDiff | 94.98 | Not recorded | 68fa77681e209131cea73147 |
| CogVideoX 5B | 94.57 | Not recorded | 68fa77681e209131cea73147 |
| CogVideoX 5B | 94.57 | Not recorded | 68fa77681e209131cea73147 |

### UDPDiff

Structural Consistency: 97.07. Result ID: 699487923a6c3148491dd14f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77681e209131cea73147.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77681e209131cea73147.md>)

Recorded evaluation configuration: {"Task ID":"depth estimation"}. This record does not assert independent reproduction.

### UDPDiff

Structural Consistency: 95.91. Result ID: 699487923a6c3148491dd154.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77681e209131cea73147.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77681e209131cea73147.md>)

Recorded evaluation configuration: {"Auxiliary Task":"depth estimation","Training Mode":"Single-task"}. This record does not assert independent reproduction.

### UDPDiff

Structural Consistency: 95.21. Result ID: 699487923a6c3148491dd14e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77681e209131cea73147.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77681e209131cea73147.md>)

Recorded evaluation configuration: {"Task ID":"segmentation"}. This record does not assert independent reproduction.

### UDPDiff

Structural Consistency: 94.98. Result ID: 699487923a6c3148491dd153.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77681e209131cea73147.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77681e209131cea73147.md>)

Recorded evaluation configuration: {"Auxiliary Task":"segmentation","Training Mode":"Single-task"}. This record does not assert independent reproduction.

### CogVideoX 5B

Structural Consistency: 94.57. Result ID: 699487923a6c3148491dd14d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77681e209131cea73147.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77a01e209131cea74741.md>)

Recorded evaluation configuration: {"Model Size":"5B"}. This record does not assert independent reproduction.

### CogVideoX 5B

Structural Consistency: 94.57. Result ID: 699487923a6c3148491dd152.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77681e209131cea73147.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77a01e209131cea74741.md>)

Recorded evaluation configuration: {"Auxiliary Task":"None","Training Mode":"Single-task"}. This record does not assert independent reproduction.

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