# Visual Reconstruction on ImageNet-1K (val)

Source: [Visual Reconstruction on ImageNet-1K (val)](<https://www.sota2.com/research/sota/visual-reconstruction-on-imagenet-1k-val>)

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

Data updated: 2026-05-26T13:40:27.832Z

Evaluation of visual reconstruction models using the validation split of the ImageNet-1K dataset, measured using metrics including rFID, SSIM, PSNR, and Codebook Usage.

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 | Visual Reconstruction |
| Dataset (including any stated split) | ImageNet-1K (val) |
| Metric | rFID |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 16 of 16 |
| 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

- [rFID](<https://www.sota2.com/research/sota/visual-reconstruction-on-imagenet-1k-val.md?metric=rFID>): Lower is better; unit: SCALAR.

- [SSIM](<https://www.sota2.com/research/sota/visual-reconstruction-on-imagenet-1k-val.md?metric=SSIM>): Higher is better; unit: SCALAR.

- [PSNR](<https://www.sota2.com/research/sota/visual-reconstruction-on-imagenet-1k-val.md?metric=PSNR>): Higher is better; unit: SCALAR.

- [Codebook Usage](<https://www.sota2.com/research/sota/visual-reconstruction-on-imagenet-1k-val.md?metric=Codebook%20Usage>): Higher is better; unit: PERCENT.

## Recorded results

| Method | rFID | Result date | Report paper |
| --- | --- | --- | --- |
| Vanilla CVQ | 0.88 | Not recorded | 6a15297fdbd7150506f7e4da |
| MoVQ | 1.05 | Not recorded | 6a15297fdbd7150506f7e4da |
| VQ w/ rotation trick | 1.1 | Not recorded | 6a15297fdbd7150506f7e4da |
| MoVQ | 1.12 | Not recorded | 6a15297fdbd7150506f7e4da |
| VQ-FC | 1.28 | Not recorded | 6a15297fdbd7150506f7e4da |
| VQ-LC | 1.29 | Not recorded | 6a15297fdbd7150506f7e4da |
| Vanilla VQ | 1.32 | Not recorded | 6a15297fdbd7150506f7e4da |
| Vanilla VQ | 1.49 | Not recorded | 6a15297fdbd7150506f7e4da |
| RVQ | 1.83 | Not recorded | 6a15297fdbd7150506f7e4da |
| Vanilla CVQ | 2.6 | Not recorded | 6a15297fdbd7150506f7e4da |
| VQ-LC | 2.62 | Not recorded | 6a15297fdbd7150506f7e4da |
| SimVQ | 2.63 | Not recorded | 6a15297fdbd7150506f7e4da |
| VQ-LC | 3.01 | Not recorded | 6a15297fdbd7150506f7e4da |
| Dynamic VQ | 3.62 | Not recorded | 6a15297fdbd7150506f7e4da |
| Dynamic VQ | 4.08 | Not recorded | 6a15297fdbd7150506f7e4da |
| Vanilla VQ | 4.84 | Not recorded | 6a15297fdbd7150506f7e4da |

### Vanilla CVQ

rFID: 0.88. Result ID: 6a15a2b4fbc373592c0ec3e0.

[Reporting paper](<https://www.sota2.com/research/paper/6a15297fdbd7150506f7e4da.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a15297fdbd7150506f7e4da.md>)

[Implementation](<https://github.com/songweii/CVQ>)

Recorded evaluation configuration: {"Token Budget":"1024 tokens","# Tokens":"1024","Ratio":"16","Codebook Size":"16,384"}. This record does not assert independent reproduction.

### MoVQ

rFID: 1.05. Result ID: 6a15a2b4fbc373592c0ec3df.

[Reporting paper](<https://www.sota2.com/research/paper/6a15297fdbd7150506f7e4da.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d741e209131cea3a058.md>)

[Implementation](<https://github.com/CompVis/taming-transformers>)

Recorded evaluation configuration: {"Token Budget":"1024 tokens","# Tokens":"16×16×4","Ratio":"16","Codebook Size":"16,384"}. This record does not assert independent reproduction.

### VQ w/ rotation trick

rFID: 1.1. Result ID: 6a15a2b4fbc373592c0ec3da.

[Reporting paper](<https://www.sota2.com/research/paper/6a15297fdbd7150506f7e4da.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa777b1e209131cea737d5.md>)

Recorded evaluation configuration: {"Token Budget":"1024 tokens","# Tokens":"32×32","Ratio":"8","Codebook Size":"16,384"}. This record does not assert independent reproduction.

### MoVQ

rFID: 1.12. Result ID: 6a15a2b4fbc373592c0ec3de.

[Reporting paper](<https://www.sota2.com/research/paper/6a15297fdbd7150506f7e4da.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d741e209131cea3a058.md>)

[Implementation](<https://github.com/CompVis/taming-transformers>)

Recorded evaluation configuration: {"Token Budget":"1024 tokens","# Tokens":"16×16×4","Ratio":"16","Codebook Size":"1,024"}. This record does not assert independent reproduction.

### VQ-FC

rFID: 1.28. Result ID: 6a15a2b4fbc373592c0ec3db.

[Reporting paper](<https://www.sota2.com/research/paper/6a15297fdbd7150506f7e4da.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6cba1e209131cea35a1f.md>)

Recorded evaluation configuration: {"Token Budget":"1024 tokens","# Tokens":"32×32","Ratio":"8","Codebook Size":"8,192"}. This record does not assert independent reproduction.

### VQ-LC

rFID: 1.29. Result ID: 6a15a2b4fbc373592c0ec3dc.

[Reporting paper](<https://www.sota2.com/research/paper/6a15297fdbd7150506f7e4da.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73a71e209131cea5db1a.md>)

[Implementation](<https://github.com/zh460045050/VQGAN-LC>)

Recorded evaluation configuration: {"Token Budget":"1024 tokens","# Tokens":"32×32","Ratio":"8","Codebook Size":"100,000"}. This record does not assert independent reproduction.

### Vanilla VQ

rFID: 1.32. Result ID: 6a15a2b4fbc373592c0ec3d9.

[Reporting paper](<https://www.sota2.com/research/paper/6a15297fdbd7150506f7e4da.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a371e209131cea26a4d.md>)

[Implementation](<https://git.io/JnyvK>)

Recorded evaluation configuration: {"Token Budget":"1024 tokens","# Tokens":"32×32","Ratio":"8","Codebook Size":"16,384","Reproduced":"true"}. This record does not assert independent reproduction.

### Vanilla VQ

rFID: 1.49. Result ID: 6a15a2b4fbc373592c0ec3d8.

[Reporting paper](<https://www.sota2.com/research/paper/6a15297fdbd7150506f7e4da.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a371e209131cea26a4d.md>)

[Implementation](<https://git.io/JnyvK>)

Recorded evaluation configuration: {"Token Budget":"1024 tokens","# Tokens":"32×32","Ratio":"8","Codebook Size":"8,192"}. This record does not assert independent reproduction.

### RVQ

rFID: 1.83. Result ID: 6a15a2b4fbc373592c0ec3dd.

[Reporting paper](<https://www.sota2.com/research/paper/6a15297fdbd7150506f7e4da.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6bf21e209131cea311a9.md>)

Recorded evaluation configuration: {"Token Budget":"1024 tokens","# Tokens":"8×8×16","Ratio":"32","Codebook Size":"16,384"}. This record does not assert independent reproduction.

### Vanilla CVQ

rFID: 2.6. Result ID: 6a15a2b4fbc373592c0ec3d7.

[Reporting paper](<https://www.sota2.com/research/paper/6a15297fdbd7150506f7e4da.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a15297fdbd7150506f7e4da.md>)

[Implementation](<https://github.com/songweii/CVQ>)

Recorded evaluation configuration: {"Token Budget":"256 tokens","# Tokens":"256","Ratio":"16","Codebook Size":"16,384"}. This record does not assert independent reproduction.

### VQ-LC

rFID: 2.62. Result ID: 6a15a2b4fbc373592c0ec3d5.

[Reporting paper](<https://www.sota2.com/research/paper/6a15297fdbd7150506f7e4da.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73a71e209131cea5db1a.md>)

[Implementation](<https://github.com/zh460045050/VQGAN-LC>)

Recorded evaluation configuration: {"Token Budget":"256 tokens","# Tokens":"16×16","Ratio":"16","Codebook Size":"100,000"}. This record does not assert independent reproduction.

### SimVQ

rFID: 2.63. Result ID: 6a15a2b4fbc373592c0ec3d6.

[Reporting paper](<https://www.sota2.com/research/paper/6a15297fdbd7150506f7e4da.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bd61e209131cea8ad53.md>)

Recorded evaluation configuration: {"Token Budget":"256 tokens","# Tokens":"16×16","Ratio":"16","Codebook Size":"16,384","Reproduced":"true"}. This record does not assert independent reproduction.

### VQ-LC

rFID: 3.01. Result ID: 6a15a2b4fbc373592c0ec3d4.

[Reporting paper](<https://www.sota2.com/research/paper/6a15297fdbd7150506f7e4da.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73a71e209131cea5db1a.md>)

[Implementation](<https://github.com/zh460045050/VQGAN-LC>)

Recorded evaluation configuration: {"Token Budget":"256 tokens","# Tokens":"16×16","Ratio":"16","Codebook Size":"16,384"}. This record does not assert independent reproduction.

### Dynamic VQ

rFID: 3.62. Result ID: 6a15a2b4fbc373592c0ec3d3.

[Reporting paper](<https://www.sota2.com/research/paper/6a15297fdbd7150506f7e4da.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f711e209131cea459b3.md>)

Recorded evaluation configuration: {"Token Budget":"256 tokens","# Tokens":"16×16","Ratio":"{8, 16, 32}","Codebook Size":"16,384","Reproduced":"true"}. This record does not assert independent reproduction.

### Dynamic VQ

rFID: 4.08. Result ID: 6a15a2b4fbc373592c0ec3d2.

[Reporting paper](<https://www.sota2.com/research/paper/6a15297fdbd7150506f7e4da.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f711e209131cea459b3.md>)

Recorded evaluation configuration: {"Token Budget":"256 tokens","# Tokens":"16×16","Ratio":"{8, 16, 32}","Codebook Size":"1,024"}. This record does not assert independent reproduction.

### Vanilla VQ

rFID: 4.84. Result ID: 6a15a2b4fbc373592c0ec3d1.

[Reporting paper](<https://www.sota2.com/research/paper/6a15297fdbd7150506f7e4da.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a371e209131cea26a4d.md>)

[Implementation](<https://git.io/JnyvK>)

Recorded evaluation configuration: {"Token Budget":"256 tokens","# Tokens":"16×16","Ratio":"16","Codebook Size":"16,384","Reproduced":"true"}. This record does not assert independent reproduction.

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