# Image Reconstruction on ImageNet-1k 256x256

Source: [Image Reconstruction on ImageNet-1k 256x256](<https://www.sota2.com/research/sota/image-reconstruction-on-imagenet-1k-256x256>)

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

Data updated: 2026-06-01T08:30:51.670Z

Evaluation of image reconstruction models using the ImageNet-1k dataset rendered at 256x256 resolution, measured using metrics such as rFID.

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 | Image Reconstruction |
| Dataset (including any stated split) | ImageNet-1k 256x256 |
| Metric | rFID |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 31 of 31 |
| 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/image-reconstruction-on-imagenet-1k-256x256.md?metric=rFID>): Lower is better; unit: SCALAR.

## Recorded results

| Method | rFID | Result date | Report paper |
| --- | --- | --- | --- |
| UniTok\* | 0.41 | Not recorded | 6a1d11e1dbd7150506f83dd6 |
| MergeTok-BL\* | 0.48 | Not recorded | 6a1d11e1dbd7150506f83dd6 |
| GigaTok-BL\* | 0.51 | Not recorded | 6a1d11e1dbd7150506f83dd6 |
| MergeTok-BL | 0.78 | Not recorded | 6a1d11e1dbd7150506f83dd6 |
| GigaTok-BL | 0.81 | Not recorded | 6a1d11e1dbd7150506f83dd6 |
| VFMTok | 0.89 | Not recorded | 6a1d11e1dbd7150506f83dd6 |
| GigaTok-SB | 0.89 | Not recorded | 6a1d11e1dbd7150506f83dd6 |
| FlowMo | 0.95 | Not recorded | 6a1d11e1dbd7150506f83dd6 |
| MergeTok-SB | 0.97 | Not recorded | 6a1d11e1dbd7150506f83dd6 |
| IBQ | 1 | Not recorded | 6a1d11e1dbd7150506f83dd6 |
| Hita | 1.03 | Not recorded | 6a1d11e1dbd7150506f83dd6 |
| Ours-VQ-L | 1.1 | Not recorded | 692689bed6b0dd3342930602 |
| Ours-VQ-XXL | 1.1 | Not recorded | 692689bed6b0dd3342930602 |
| OmniTokenizer | 1.11 | Not recorded | 6a1d11e1dbd7150506f83dd6 |
| MergeVQ-GR | 1.12 | Not recorded | 6a1d11e1dbd7150506f83dd6 |
| VFMTok-L | 1.13 | Not recorded | 692689bed6b0dd3342930602 |
| VAEVQ | 1.14 | Not recorded | 6a1d11e1dbd7150506f83dd6 |
| OpenMAGVIT2 | 1.17 | Not recorded | 6a1d11e1dbd7150506f83dd6 |
| ViT-VQGAN | 1.28 | Not recorded | 6a1d11e1dbd7150506f83dd6 |
| MergeVQ-GR | 1.48 | Not recorded | 6a1d11e1dbd7150506f83dd6 |
| Titok-B-64 | 1.7 | Not recorded | 6a1d11e1dbd7150506f83dd6 |
| Titok-S-128 | 1.71 | Not recorded | 6a1d11e1dbd7150506f83dd6 |
| LlamaGen | 2.19 | Not recorded | 6a1d11e1dbd7150506f83dd6 |
| LlamaGen-L | 2.21 | Not recorded | 692689bed6b0dd3342930602 |
| Titok-L-32 | 2.21 | Not recorded | 6a1d11e1dbd7150506f83dd6 |
| MaskGIT | 2.28 | Not recorded | 692689bed6b0dd3342930602 |
| RQ-Transformer | 3.2 | Not recorded | 692689bed6b0dd3342930602 |
| LlamaGen-L | 4.98 | Not recorded | 692689bed6b0dd3342930602 |
| VQGAN | 4.98 | Not recorded | 692689bed6b0dd3342930602 |
| LlamaGen-XXL | 4.98 | Not recorded | 692689bed6b0dd3342930602 |
| Taming-VQGAN | 7.94 | Not recorded | 6a1d11e1dbd7150506f83dd6 |

### UniTok\*

rFID: 0.41. Result ID: 6a1d4329fbc373592c1674b3.

[Reporting paper](<https://www.sota2.com/research/paper/6a1d11e1dbd7150506f83dd6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78951e209131cea791bb.md>)

Recorded evaluation configuration: {"Ratio":"16","VQ Codebook Type":"2D MSQ","VQ Codebook #Tok.":"16^2","VQ Codebook #Code":"2^15","DINO discriminator":"frozen"}. This record does not assert independent reproduction.

### MergeTok-BL\*

rFID: 0.48. Result ID: 6a1d4329fbc373592c1674c2.

[Reporting paper](<https://www.sota2.com/research/paper/6a1d11e1dbd7150506f83dd6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a1d11e1dbd7150506f83dd6.md>)

Recorded evaluation configuration: {"Ratio":"16","VQ Codebook Type":"1D VAE+VQ","VQ Codebook #Tok.":"256","VQ Codebook #Code":"2^14","DINO discriminator":"frozen"}. This record does not assert independent reproduction.

### GigaTok-BL\*

rFID: 0.51. Result ID: 6a1d4329fbc373592c1674bd.

[Reporting paper](<https://www.sota2.com/research/paper/6a1d11e1dbd7150506f83dd6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7acd1e209131cea8530d.md>)

[Implementation](<https://silentview.github.io/GigaTok/>)

Recorded evaluation configuration: {"Ratio":"16","VQ Codebook Type":"1D VQ","VQ Codebook #Tok.":"256","VQ Codebook #Code":"2^14","DINO discriminator":"frozen"}. This record does not assert independent reproduction.

### MergeTok-BL

rFID: 0.78. Result ID: 6a1d4329fbc373592c1674c3.

[Reporting paper](<https://www.sota2.com/research/paper/6a1d11e1dbd7150506f83dd6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a1d11e1dbd7150506f83dd6.md>)

Recorded evaluation configuration: {"Ratio":"16","VQ Codebook Type":"1D VAE+VQ","VQ Codebook #Tok.":"256","VQ Codebook #Code":"2^14"}. This record does not assert independent reproduction.

### GigaTok-BL

rFID: 0.81. Result ID: 6a1d4329fbc373592c1674be.

[Reporting paper](<https://www.sota2.com/research/paper/6a1d11e1dbd7150506f83dd6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7acd1e209131cea8530d.md>)

[Implementation](<https://silentview.github.io/GigaTok/>)

Recorded evaluation configuration: {"Ratio":"16","VQ Codebook Type":"1D VQ","VQ Codebook #Tok.":"256","VQ Codebook #Code":"2^14"}. This record does not assert independent reproduction.

### VFMTok

rFID: 0.89. Result ID: 6a1d4329fbc373592c1674b2.

[Reporting paper](<https://www.sota2.com/research/paper/6a1d11e1dbd7150506f83dd6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79fb1e209131cea806ef.md>)

Recorded evaluation configuration: {"Ratio":"16","VQ Codebook Type":"2D VQ","VQ Codebook #Tok.":"16^2","VQ Codebook #Code":"2^14"}. This record does not assert independent reproduction.

### GigaTok-SB

rFID: 0.89. Result ID: 6a1d4329fbc373592c1674bc.

[Reporting paper](<https://www.sota2.com/research/paper/6a1d11e1dbd7150506f83dd6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7acd1e209131cea8530d.md>)

[Implementation](<https://silentview.github.io/GigaTok/>)

Recorded evaluation configuration: {"Ratio":"16","VQ Codebook Type":"1D VQ","VQ Codebook #Tok.":"256","VQ Codebook #Code":"2^14"}. This record does not assert independent reproduction.

### FlowMo

rFID: 0.95. Result ID: 6a1d4329fbc373592c1674b6.

[Reporting paper](<https://www.sota2.com/research/paper/6a1d11e1dbd7150506f83dd6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77c91e209131cea75579.md>)

Recorded evaluation configuration: {"Ratio":"16","VQ Codebook Type":"2D Diff.+LFQ","VQ Codebook #Tok.":"16^2","VQ Codebook #Code":"2^18"}. This record does not assert independent reproduction.

### MergeTok-SB

rFID: 0.97. Result ID: 6a1d4329fbc373592c1674c1.

[Reporting paper](<https://www.sota2.com/research/paper/6a1d11e1dbd7150506f83dd6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a1d11e1dbd7150506f83dd6.md>)

Recorded evaluation configuration: {"Ratio":"16","VQ Codebook Type":"1D VAE+VQ","VQ Codebook #Tok.":"256","VQ Codebook #Code":"2^14"}. This record does not assert independent reproduction.

### IBQ

rFID: 1. Result ID: 6a1d4329fbc373592c1674b5.

[Reporting paper](<https://www.sota2.com/research/paper/6a1d11e1dbd7150506f83dd6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa775d1e209131cea72ba7.md>)

[Implementation](<https://github.com/TencentARC/SEED-Voken>)

Recorded evaluation configuration: {"Ratio":"16","VQ Codebook Type":"2D LFQ","VQ Codebook #Tok.":"16^2","VQ Codebook #Code":"2^18"}. This record does not assert independent reproduction.

### Hita

rFID: 1.03. Result ID: 6a1d4329fbc373592c1674bf.

[Reporting paper](<https://www.sota2.com/research/paper/6a1d11e1dbd7150506f83dd6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79f91e209131cea80635.md>)

Recorded evaluation configuration: {"Ratio":"16","VQ Codebook Type":"1D VQ","VQ Codebook #Tok.":"569","VQ Codebook #Code":"2^14"}. This record does not assert independent reproduction.

### Ours-VQ-L

rFID: 1.1. Result ID: 69c3e51447f34ce924ad75f8.

[Reporting paper](<https://www.sota2.com/research/paper/692689bed6b0dd3342930602.md>)

[Method origin paper](<https://www.sota2.com/research/paper/692689bed6b0dd3342930602.md>)

Recorded evaluation configuration: {"Tokenizer":"DINO-Tok-VQ"}. This record does not assert independent reproduction.

### Ours-VQ-XXL

rFID: 1.1. Result ID: 69c3e51447f34ce924ad75fc.

[Reporting paper](<https://www.sota2.com/research/paper/692689bed6b0dd3342930602.md>)

[Method origin paper](<https://www.sota2.com/research/paper/692689bed6b0dd3342930602.md>)

Recorded evaluation configuration: {"Tokenizer":"DINO-Tok-VQ"}. This record does not assert independent reproduction.

### OmniTokenizer

rFID: 1.11. Result ID: 6a1d4329fbc373592c1674b1.

[Reporting paper](<https://www.sota2.com/research/paper/6a1d11e1dbd7150506f83dd6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73961e209131cea5d652.md>)

Recorded evaluation configuration: {"Ratio":"16","VQ Codebook Type":"2D VQ+VAE","VQ Codebook #Tok.":"16^2","VQ Codebook #Code":"2^13"}. This record does not assert independent reproduction.

### MergeVQ-GR

rFID: 1.12. Result ID: 6a1d4329fbc373592c1674bb.

[Reporting paper](<https://www.sota2.com/research/paper/6a1d11e1dbd7150506f83dd6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77b41e209131cea74dd6.md>)

Recorded evaluation configuration: {"Ratio":"16","VQ Codebook Type":"1D LFQ","VQ Codebook #Tok.":"256","VQ Codebook #Code":"2^18"}. This record does not assert independent reproduction.

### VFMTok-L

rFID: 1.13. Result ID: 69c3e51447f34ce924ad75f7.

[Reporting paper](<https://www.sota2.com/research/paper/692689bed6b0dd3342930602.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79fb1e209131cea806ef.md>)

Recorded evaluation configuration: {"Tokenizer":"VFMTok"}. This record does not assert independent reproduction.

### VAEVQ

rFID: 1.14. Result ID: 6a1d4329fbc373592c1674c0.

[Reporting paper](<https://www.sota2.com/research/paper/6a1d11e1dbd7150506f83dd6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6912c3492828b7cc6937e094.md>)

Recorded evaluation configuration: {"Ratio":"16","VQ Codebook Type":"VAE+VQ","VQ Codebook #Tok.":"256","VQ Codebook #Code":"2^14"}. This record does not assert independent reproduction.

### OpenMAGVIT2

rFID: 1.17. Result ID: 6a1d4329fbc373592c1674b4.

[Reporting paper](<https://www.sota2.com/research/paper/6a1d11e1dbd7150506f83dd6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76e01e209131cea6f87d.md>)

[Implementation](<https://github.com/TencentARC/Open-MAGVIT2>)

Recorded evaluation configuration: {"Ratio":"16","VQ Codebook Type":"2D LFQ","VQ Codebook #Tok.":"16^2","VQ Codebook #Code":"2^18"}. This record does not assert independent reproduction.

### ViT-VQGAN

rFID: 1.28. Result ID: 6a1d4329fbc373592c1674af.

[Reporting paper](<https://www.sota2.com/research/paper/6a1d11e1dbd7150506f83dd6.md>)

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

Recorded evaluation configuration: {"Ratio":"8","VQ Codebook Type":"2D VQ","VQ Codebook #Tok.":"32^2","VQ Codebook #Code":"2^13"}. This record does not assert independent reproduction.

### MergeVQ-GR

rFID: 1.48. Result ID: 6a1d4329fbc373592c1674ba.

[Reporting paper](<https://www.sota2.com/research/paper/6a1d11e1dbd7150506f83dd6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77b41e209131cea74dd6.md>)

Recorded evaluation configuration: {"Ratio":"16","VQ Codebook Type":"1D LFQ","VQ Codebook #Tok.":"144","VQ Codebook #Code":"2^18"}. This record does not assert independent reproduction.

### Titok-B-64

rFID: 1.7. Result ID: 6a1d4329fbc373592c1674b8.

[Reporting paper](<https://www.sota2.com/research/paper/6a1d11e1dbd7150506f83dd6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73991e209131cea5d7e1.md>)

[Implementation](<https://yucornetto.github.io/projects/titok.html>)

Recorded evaluation configuration: {"Ratio":"16","VQ Codebook Type":"1D VQ","VQ Codebook #Tok.":"64","VQ Codebook #Code":"2^12"}. This record does not assert independent reproduction.

### Titok-S-128

rFID: 1.71. Result ID: 6a1d4329fbc373592c1674b7.

[Reporting paper](<https://www.sota2.com/research/paper/6a1d11e1dbd7150506f83dd6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73991e209131cea5d7e1.md>)

[Implementation](<https://yucornetto.github.io/projects/titok.html>)

Recorded evaluation configuration: {"Ratio":"16","VQ Codebook Type":"1D VQ","VQ Codebook #Tok.":"128","VQ Codebook #Code":"2^12"}. This record does not assert independent reproduction.

### LlamaGen

rFID: 2.19. Result ID: 6a1d4329fbc373592c1674b0.

[Reporting paper](<https://www.sota2.com/research/paper/6a1d11e1dbd7150506f83dd6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa738a1e209131cea5d111.md>)

[Implementation](<https://github.com/FoundationVision/LlamaGen>)

Recorded evaluation configuration: {"Ratio":"16","VQ Codebook Type":"2D VQ","VQ Codebook #Tok.":"16^2","VQ Codebook #Code":"2^14"}. This record does not assert independent reproduction.

### LlamaGen-L

rFID: 2.21. Result ID: 69c3e51447f34ce924ad75f5.

[Reporting paper](<https://www.sota2.com/research/paper/692689bed6b0dd3342930602.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa738a1e209131cea5d111.md>)

[Implementation](<https://github.com/FoundationVision/LlamaGen>)

Recorded evaluation configuration: {"Tokenizer":"TiTok-L \[54\]"}. This record does not assert independent reproduction.

### Titok-L-32

rFID: 2.21. Result ID: 6a1d4329fbc373592c1674b9.

[Reporting paper](<https://www.sota2.com/research/paper/6a1d11e1dbd7150506f83dd6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73991e209131cea5d7e1.md>)

[Implementation](<https://yucornetto.github.io/projects/titok.html>)

Recorded evaluation configuration: {"Ratio":"16","VQ Codebook Type":"1D VQ","VQ Codebook #Tok.":"32","VQ Codebook #Code":"2^12"}. This record does not assert independent reproduction.

### MaskGIT

rFID: 2.28. Result ID: 69c3e51447f34ce924ad75f6.

[Reporting paper](<https://www.sota2.com/research/paper/692689bed6b0dd3342930602.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6bc61e209131cea303c7.md>)

Recorded evaluation configuration: {"Tokenizer":"MaskGiT \[8\]"}. This record does not assert independent reproduction.

### RQ-Transformer

rFID: 3.2. Result ID: 69c3e51447f34ce924ad75fb.

[Reporting paper](<https://www.sota2.com/research/paper/692689bed6b0dd3342930602.md>)

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

Recorded evaluation configuration: {"Tokenizer":"RQ-VAE \[22\]"}. This record does not assert independent reproduction.

### LlamaGen-L

rFID: 4.98. Result ID: 69c3e51447f34ce924ad75f4.

[Reporting paper](<https://www.sota2.com/research/paper/692689bed6b0dd3342930602.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa738a1e209131cea5d111.md>)

[Implementation](<https://github.com/FoundationVision/LlamaGen>)

Recorded evaluation configuration: {"Tokenizer":"VQGAN \[14\]"}. This record does not assert independent reproduction.

### VQGAN

rFID: 4.98. Result ID: 69c3e51447f34ce924ad75f9.

[Reporting paper](<https://www.sota2.com/research/paper/692689bed6b0dd3342930602.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa64691e209131cea048db.md>)

Recorded evaluation configuration: {"Tokenizer":"VQGAN \[14\]"}. This record does not assert independent reproduction.

### LlamaGen-XXL

rFID: 4.98. Result ID: 69c3e51447f34ce924ad75fa.

[Reporting paper](<https://www.sota2.com/research/paper/692689bed6b0dd3342930602.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa738a1e209131cea5d111.md>)

[Implementation](<https://github.com/FoundationVision/LlamaGen>)

Recorded evaluation configuration: {"Tokenizer":"VQGAN \[14\]"}. This record does not assert independent reproduction.

### Taming-VQGAN

rFID: 7.94. Result ID: 6a1d4329fbc373592c1674ae.

[Reporting paper](<https://www.sota2.com/research/paper/6a1d11e1dbd7150506f83dd6.md>)

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

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

Recorded evaluation configuration: {"Ratio":"16","VQ Codebook Type":"2D VQ","VQ Codebook #Tok.":"16^2","VQ Codebook #Code":"2^10"}. This record does not assert independent reproduction.

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