# Video Prediction on BAIR Robot Pushing

Source: [Video Prediction on BAIR Robot Pushing](<https://www.sota2.com/research/sota/video-prediction-on-bair-robot-pushing>)

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

Data updated: 2026-02-26T05:45:05.511Z

Evaluation of video prediction performance on the BAIR Robot Pushing dataset, typically measured using metrics like 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 prediction |
| Dataset (including any stated split) | BAIR Robot Pushing |
| Metric | FVD |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 38 of 38 |
| 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

- [FVD](<https://www.sota2.com/research/sota/video-prediction-on-bair-robot-pushing.md?metric=FVD>): Lower is better; unit: SCALAR.

- [PSNR](<https://www.sota2.com/research/sota/video-prediction-on-bair-robot-pushing.md?metric=PSNR>): Higher is better; unit: SCALAR.

- [SSIM](<https://www.sota2.com/research/sota/video-prediction-on-bair-robot-pushing.md?metric=SSIM>): Higher is better; unit: SCALAR.

- [LPIPS](<https://www.sota2.com/research/sota/video-prediction-on-bair-robot-pushing.md?metric=LPIPS>): Lower is better; unit: SCALAR.

## Recorded results

| Method | FVD | Result date | Report paper |
| --- | --- | --- | --- |
| MAGVIT-L-FP | 31 | Not recorded | 68fa6eee1e209131cea42f71 |
| MAGVIT-B-FP | 48 | Not recorded | 68fa6eee1e209131cea42f71 |
| iVideoGPT | 60.8 | Not recorded | 68fa757b1e209131cea683e5 |
| MAGVIT-L-FP | 62 | Not recorded | 68fa6eee1e209131cea42f71 |
| MAGVIT | 62 | Not recorded | 68fa757b1e209131cea683e5 |
| Video Diffusion | 66.92 | Not recorded | 68fa6cd51e209131cea36149 |
| Video Diffusion | 68.19 | Not recorded | 68fa6cd51e209131cea36149 |
| MaskViT | 70.5 | Not recorded | 68fa757b1e209131cea683e5 |
| iVideoGPT | 75 | Not recorded | 68fa757b1e209131cea683e5 |
| MAGVIT-B-FP | 76 | Not recorded | 68fa6eee1e209131cea42f71 |
| RaMViD | 84 | Not recorded | 68fa6eee1e209131cea42f71 |
| NUWA | 86.9 | Not recorded | 68fa6cd51e209131cea36149 |
| NÜWA | 87 | Not recorded | 68fa6eee1e209131cea42f71 |
| MCVD | 89.5 | Not recorded | 68fa757b1e209131cea683e5 |
| MCVD | 90 | Not recorded | 68fa6eee1e209131cea42f71 |
| FitVid | 93.6 | Not recorded | 68fa6cd51e209131cea36149 |
| FitVid | 93.6 | Not recorded | 68fa757b1e209131cea683e5 |
| MaskViT | 93.7 | Not recorded | 68fa757b1e209131cea683e5 |
| VideoTransformer | 94 | Not recorded | 68fa6cd51e209131cea36149 |
| MaskViT | 94 | Not recorded | 68fa6eee1e209131cea42f71 |
| FitVid | 94 | Not recorded | 68fa6eee1e209131cea42f71 |
| Phenaki | 97 | Not recorded | 68fa6eee1e209131cea42f71 |
| CCVS | 99 | Not recorded | 68fa6cd51e209131cea36149 |
| CCVS | 99 | Not recorded | 68fa6eee1e209131cea42f71 |
| Transframer | 100 | Not recorded | 68fa6cd51e209131cea36149 |
| Transframer | 100 | Not recorded | 68fa6eee1e209131cea42f71 |
| TrIVD-GAN-FP | 103 | Not recorded | 68fa6eee1e209131cea42f71 |
| VideoGPT | 103.3 | Not recorded | 68fa6cd51e209131cea36149 |
| TrIVD-GAN-FP | 103.3 | Not recorded | 68fa6cd51e209131cea36149 |
| VideoGPT | 103.3 | Not recorded | 68fa757b1e209131cea683e5 |
| VideoGPT | 103.3 | Not recorded | 68fa6ad11e209131cea2a82d |
| TrIVD-GAN-FP | 103.3 | Not recorded | 68fa6ad11e209131cea2a82d |
| DVD-GAN | 109.8 | Not recorded | 68fa6cd51e209131cea36149 |
| DVD-GAN-FP | 109.8 | Not recorded | 68fa6ad11e209131cea2a82d |
| SAVP | 116.4 | Not recorded | 68fa6ad11e209131cea2a82d |
| LVT | 125.8 | Not recorded | 68fa6ad11e209131cea2a82d |
| SV2P | 262.5 | Not recorded | 68fa6ad11e209131cea2a82d |
| Video Transformer | 942 | Not recorded | 68fa6ad11e209131cea2a82d |

### MAGVIT-L-FP

FVD: 31. Result ID: 699072373a6c314849009e7c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6eee1e209131cea42f71.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eee1e209131cea42f71.md>)

[Implementation](<https://magvit.cs.cmu.edu/>)

Recorded evaluation configuration: {"Model Size":"464M","Evaluation Protocol":"Debiased"}. This record does not assert independent reproduction.

### MAGVIT-B-FP

FVD: 48. Result ID: 699072373a6c314849009e7a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6eee1e209131cea42f71.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eee1e209131cea42f71.md>)

[Implementation](<https://magvit.cs.cmu.edu/>)

Recorded evaluation configuration: {"Model Size":"128M","Evaluation Protocol":"Debiased"}. This record does not assert independent reproduction.

### iVideoGPT

FVD: 60.8. Result ID: 6993b68a3a6c3148491803ad.

[Reporting paper](<https://www.sota2.com/research/paper/68fa757b1e209131cea683e5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa757b1e209131cea683e5.md>)

[Implementation](<https://thuml.github.io/iVideoGPT>)

Recorded evaluation configuration: {"Conditioning":"action-conditioned","Resolution":"64x64"}. This record does not assert independent reproduction.

### MAGVIT-L-FP

FVD: 62. Result ID: 699072373a6c314849009e7b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6eee1e209131cea42f71.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eee1e209131cea42f71.md>)

[Implementation](<https://magvit.cs.cmu.edu/>)

Recorded evaluation configuration: {"Model Size":"464M","Evaluation Protocol":"Standard"}. This record does not assert independent reproduction.

### MAGVIT

FVD: 62. Result ID: 6993b68a3a6c3148491803aa.

[Reporting paper](<https://www.sota2.com/research/paper/68fa757b1e209131cea683e5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eee1e209131cea42f71.md>)

[Implementation](<https://magvit.cs.cmu.edu/>)

Recorded evaluation configuration: {"Conditioning":"action-free","Resolution":"64x64"}. This record does not assert independent reproduction.

### Video Diffusion

FVD: 66.92. Result ID: 698f41853a6c314849f7805c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6cd51e209131cea36149.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6cd51e209131cea36149.md>)

[Implementation](<https://video-diffusion.github.io/>)

Recorded evaluation configuration: {"sampler":"Langevin","steps":256}. This record does not assert independent reproduction.

### Video Diffusion

FVD: 68.19. Result ID: 698f41853a6c314849f7805b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6cd51e209131cea36149.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6cd51e209131cea36149.md>)

[Implementation](<https://video-diffusion.github.io/>)

Recorded evaluation configuration: {"sampler":"ancestral","steps":512}. This record does not assert independent reproduction.

### MaskViT

FVD: 70.5. Result ID: 6993b68a3a6c3148491803ac.

[Reporting paper](<https://www.sota2.com/research/paper/68fa757b1e209131cea683e5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d351e209131cea3871a.md>)

Recorded evaluation configuration: {"Conditioning":"action-conditioned","Resolution":"64x64"}. This record does not assert independent reproduction.

### iVideoGPT

FVD: 75. Result ID: 6993b68a3a6c3148491803ab.

[Reporting paper](<https://www.sota2.com/research/paper/68fa757b1e209131cea683e5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa757b1e209131cea683e5.md>)

[Implementation](<https://thuml.github.io/iVideoGPT>)

Recorded evaluation configuration: {"Conditioning":"action-free","Resolution":"64x64"}. This record does not assert independent reproduction.

### MAGVIT-B-FP

FVD: 76. Result ID: 699072373a6c314849009e79.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6eee1e209131cea42f71.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eee1e209131cea42f71.md>)

[Implementation](<https://magvit.cs.cmu.edu/>)

Recorded evaluation configuration: {"Model Size":"128M","Evaluation Protocol":"Standard"}. This record does not assert independent reproduction.

### RaMViD

FVD: 84. Result ID: 699072373a6c314849009e78.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6eee1e209131cea42f71.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6dea1e209131cea3ce5f.md>)

[Implementation](<https://github.com/Tobi-r9/RaMViD>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### NUWA

FVD: 86.9. Result ID: 698f41853a6c314849f7805a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6cd51e209131cea36149.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b4e1e209131cea2d7e7.md>)

[Implementation](<https://github.com/microsoft/NUWA>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### NÜWA

FVD: 87. Result ID: 699072373a6c314849009e77.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6eee1e209131cea42f71.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### MCVD

FVD: 89.5. Result ID: 6993b68a3a6c3148491803a9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa757b1e209131cea683e5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d9f1e209131cea3aeca.md>)

[Implementation](<https://github.com/voletiv/mcvd-pytorch>)

Recorded evaluation configuration: {"Conditioning":"action-free","Resolution":"64x64"}. This record does not assert independent reproduction.

### MCVD

FVD: 90. Result ID: 699072373a6c314849009e76.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6eee1e209131cea42f71.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d9f1e209131cea3aeca.md>)

[Implementation](<https://github.com/voletiv/mcvd-pytorch>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### FitVid

FVD: 93.6. Result ID: 698f41853a6c314849f78059.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6cd51e209131cea36149.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a391e209131cea26bde.md>)

[Implementation](<https://github.com/google-research/fitvid>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### FitVid

FVD: 93.6. Result ID: 6993b68a3a6c3148491803a8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa757b1e209131cea683e5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a391e209131cea26bde.md>)

[Implementation](<https://github.com/google-research/fitvid>)

Recorded evaluation configuration: {"Conditioning":"action-free","Resolution":"64x64"}. This record does not assert independent reproduction.

### MaskViT

FVD: 93.7. Result ID: 6993b68a3a6c3148491803a7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa757b1e209131cea683e5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d351e209131cea3871a.md>)

Recorded evaluation configuration: {"Conditioning":"action-free","Resolution":"64x64"}. This record does not assert independent reproduction.

### VideoTransformer

FVD: 94. Result ID: 698f41853a6c314849f78058.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6cd51e209131cea36149.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67061e209131cea13e6e.md>)

[Implementation](<https://bit.ly/2Zb017f>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### MaskViT

FVD: 94. Result ID: 699072373a6c314849009e74.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6eee1e209131cea42f71.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d351e209131cea3871a.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### FitVid

FVD: 94. Result ID: 699072373a6c314849009e75.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6eee1e209131cea42f71.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a391e209131cea26bde.md>)

[Implementation](<https://github.com/google-research/fitvid>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Phenaki

FVD: 97. Result ID: 699072373a6c314849009e71.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6eee1e209131cea42f71.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d8b1e209131cea3a7e8.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### CCVS

FVD: 99. Result ID: 698f41853a6c314849f78057.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6cd51e209131cea36149.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b1c1e209131cea2c47f.md>)

[Implementation](<https://16lemoing.github.io/ccvs>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### CCVS

FVD: 99. Result ID: 699072373a6c314849009e70.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6eee1e209131cea42f71.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b1c1e209131cea2c47f.md>)

[Implementation](<https://16lemoing.github.io/ccvs>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Transframer

FVD: 100. Result ID: 698f41853a6c314849f78056.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6cd51e209131cea36149.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c781e209131cea3421e.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Transframer

FVD: 100. Result ID: 699072373a6c314849009e73.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6eee1e209131cea42f71.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c781e209131cea3421e.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### TrIVD-GAN-FP

FVD: 103. Result ID: 699072373a6c314849009e72.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6eee1e209131cea42f71.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b491e209131cea2d4b9.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### VideoGPT

FVD: 103.3. Result ID: 698f41853a6c314849f78054.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6cd51e209131cea36149.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ad11e209131cea2a82d.md>)

[Implementation](<https://wilson1yan.github.io/videogpt/index.html>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### TrIVD-GAN-FP

FVD: 103.3. Result ID: 698f41853a6c314849f78055.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6cd51e209131cea36149.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b491e209131cea2d4b9.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### VideoGPT

FVD: 103.3. Result ID: 6993b68a3a6c3148491803a6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa757b1e209131cea683e5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ad11e209131cea2a82d.md>)

[Implementation](<https://wilson1yan.github.io/videogpt/index.html>)

Recorded evaluation configuration: {"Conditioning":"action-free","Resolution":"64x64"}. This record does not assert independent reproduction.

### VideoGPT

FVD: 103.3. Result ID: 699ad4cf5698665ec4640954.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ad11e209131cea2a82d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ad11e209131cea2a82d.md>)

[Implementation](<https://wilson1yan.github.io/videogpt/index.html>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### TrIVD-GAN-FP

FVD: 103.3. Result ID: 699ad4cf5698665ec4640955.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ad11e209131cea2a82d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b491e209131cea2d4b9.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### DVD-GAN

FVD: 109.8. Result ID: 698f41853a6c314849f78053.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6cd51e209131cea36149.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa667a1e209131cea10729.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### DVD-GAN-FP

FVD: 109.8. Result ID: 699ad4cf5698665ec4640953.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ad11e209131cea2a82d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa667a1e209131cea10729.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### SAVP

FVD: 116.4. Result ID: 699ad4cf5698665ec4640952.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ad11e209131cea2a82d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa642d1e209131cea0326e.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### LVT

FVD: 125.8. Result ID: 699ad4cf5698665ec4640951.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ad11e209131cea2a82d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67cc1e209131cea187b1.md>)

[Implementation](<https://github.com/rakhimovv/lvt>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### SV2P

FVD: 262.5. Result ID: 699ad4cf5698665ec4640950.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ad11e209131cea2a82d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa64261e209131cea02e93.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Video Transformer

FVD: 942. Result ID: 699ad4cf5698665ec4640956.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ad11e209131cea2a82d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67061e209131cea13e6e.md>)

[Implementation](<https://bit.ly/2Zb017f>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

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