# Class Conditional Image Generation on ImageNet 256x256 1k (train val)

Source: [Class Conditional Image Generation on ImageNet 256x256 1k (train val)](<https://www.sota2.com/research/sota/class-conditional-image-generation-on-imagenet-256x256-1k-train-val>)

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

Data updated: 2026-02-26T05:44:25.160Z

Evaluation of class-conditional image generation models on the 1k subset of ImageNet rendered at 256x256 resolution, using both training and validation splits.

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 | Class Conditional Image Generation |
| Dataset (including any stated split) | ImageNet 256x256 1k (train val) |
| Metric | FID |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 17 of 17 |
| 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

- [FID](<https://www.sota2.com/research/sota/class-conditional-image-generation-on-imagenet-256x256-1k-train-val.md?metric=FID>): Lower is better; unit: SCALAR.

- [sFID](<https://www.sota2.com/research/sota/class-conditional-image-generation-on-imagenet-256x256-1k-train-val.md?metric=sFID>): Lower is better; unit: SCALAR.

- [IS](<https://www.sota2.com/research/sota/class-conditional-image-generation-on-imagenet-256x256-1k-train-val.md?metric=IS>): Higher is better; unit: SCALAR.

- [Precision](<https://www.sota2.com/research/sota/class-conditional-image-generation-on-imagenet-256x256-1k-train-val.md?metric=Precision>): Higher is better; unit: PERCENT.

- [Recall](<https://www.sota2.com/research/sota/class-conditional-image-generation-on-imagenet-256x256-1k-train-val.md?metric=Recall>): Higher is better; unit: PERCENT.

## Recorded results

| Method | FID | Result date | Report paper |
| --- | --- | --- | --- |
| DiT-XL/2-G | 2.27 | Not recorded | 68fa715b1e209131cea50d2a |
| DIFFUSSM-XL-G | 2.28 | Not recorded | 68fa715b1e209131cea50d2a |
| U-ViT-H/2-G | 2.29 | Not recorded | 68fa715b1e209131cea50d2a |
| StyleGAN-XL | 2.3 | Not recorded | 68fa715b1e209131cea50d2a |
| ADM-G, ADM-U | 3.6 | Not recorded | 68fa715b1e209131cea50d2a |
| LDM-4-G | 3.95 | Not recorded | 68fa715b1e209131cea50d2a |
| ADM-G | 4.59 | Not recorded | 68fa715b1e209131cea50d2a |
| CDM | 4.88 | Not recorded | 68fa715b1e209131cea50d2a |
| MaskGIT | 6.18 | Not recorded | 68fa715b1e209131cea50d2a |
| BigGAN-deep | 6.95 | Not recorded | 68fa715b1e209131cea50d2a |
| ADM-U | 7.49 | Not recorded | 68fa715b1e209131cea50d2a |
| LDM-8-G | 7.76 | Not recorded | 68fa715b1e209131cea50d2a |
| DIFFUSSM-XL | 9.07 | Not recorded | 68fa715b1e209131cea50d2a |
| DiT-XL/2 | 9.62 | Not recorded | 68fa715b1e209131cea50d2a |
| LDM-4 | 10.56 | Not recorded | 68fa715b1e209131cea50d2a |
| ADM | 10.94 | Not recorded | 68fa715b1e209131cea50d2a |
| LDM-8 | 15.51 | Not recorded | 68fa715b1e209131cea50d2a |

### DiT-XL/2-G

FID: 2.27. Result ID: 6991ad693a6c31484909d778.

[Reporting paper](<https://www.sota2.com/research/paper/68fa715b1e209131cea50d2a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ea91e209131cea41328.md>)

Recorded evaluation configuration: {"Total Images(M)":1792,"Total Gflops":"2.13 x 10^11","Classifier-free Guidance":true}. This record does not assert independent reproduction.

### DIFFUSSM-XL-G

FID: 2.28. Result ID: 6991ad693a6c31484909d779.

[Reporting paper](<https://www.sota2.com/research/paper/68fa715b1e209131cea50d2a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa715b1e209131cea50d2a.md>)

Recorded evaluation configuration: {"Total Images(M)":660,"Total Gflops":"1.85 x 10^11","Classifier-free Guidance":true}. This record does not assert independent reproduction.

### U-ViT-H/2-G

FID: 2.29. Result ID: 6991ad693a6c31484909d777.

[Reporting paper](<https://www.sota2.com/research/paper/68fa715b1e209131cea50d2a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ee71e209131cea42caa.md>)

[Implementation](<https://github.com/baofff/U-ViT>)

Recorded evaluation configuration: {"Total Images(M)":512,"Total Gflops":"6.81 x 10^10","Classifier-free Guidance":true}. This record does not assert independent reproduction.

### StyleGAN-XL

FID: 2.3. Result ID: 6991ad693a6c31484909d76b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa715b1e209131cea50d2a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c741e209131cea34012.md>)

Recorded evaluation configuration: {"Classifier-free Guidance":false}. This record does not assert independent reproduction.

### ADM-G, ADM-U

FID: 3.6. Result ID: 6991ad693a6c31484909d774.

[Reporting paper](<https://www.sota2.com/research/paper/68fa715b1e209131cea50d2a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a0e1e209131cea25dba.md>)

[Implementation](<https://github.com/openai/guided-diffusion>)

Recorded evaluation configuration: {"Total Images(M)":507,"Total Gflops":"3.76 x 10^12","Classifier-free Guidance":true}. This record does not assert independent reproduction.

### LDM-4-G

FID: 3.95. Result ID: 6991ad693a6c31484909d776.

[Reporting paper](<https://www.sota2.com/research/paper/68fa715b1e209131cea50d2a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c391e209131cea32e4a.md>)

[Implementation](<https://github.com/CompVis/latent-diffusion>)

Recorded evaluation configuration: {"Total Images(M)":213,"Total Gflops":"2.22 x 10^10","Classifier-free Guidance":true}. This record does not assert independent reproduction.

### ADM-G

FID: 4.59. Result ID: 6991ad693a6c31484909d773.

[Reporting paper](<https://www.sota2.com/research/paper/68fa715b1e209131cea50d2a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a0e1e209131cea25dba.md>)

[Implementation](<https://github.com/openai/guided-diffusion>)

Recorded evaluation configuration: {"Total Images(M)":507,"Total Gflops":"5.68 x 10^11","Classifier-free Guidance":true}. This record does not assert independent reproduction.

### CDM

FID: 4.88. Result ID: 6991ad693a6c31484909d76e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa715b1e209131cea50d2a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b781e209131cea2e5b5.md>)

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

Recorded evaluation configuration: {"Classifier-free Guidance":false}. This record does not assert independent reproduction.

### MaskGIT

FID: 6.18. Result ID: 6991ad693a6c31484909d76a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa715b1e209131cea50d2a.md>)

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

Recorded evaluation configuration: {"Total Images(M)":355,"Classifier-free Guidance":false}. This record does not assert independent reproduction.

### BigGAN-deep

FID: 6.95. Result ID: 6991ad693a6c31484909d769.

[Reporting paper](<https://www.sota2.com/research/paper/68fa715b1e209131cea50d2a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65661e209131cea0a090.md>)

[Implementation](<https://tfhub.dev/s?q=biggan>)

Recorded evaluation configuration: {"Classifier-free Guidance":false}. This record does not assert independent reproduction.

### ADM-U

FID: 7.49. Result ID: 6991ad693a6c31484909d76d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa715b1e209131cea50d2a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a0e1e209131cea25dba.md>)

[Implementation](<https://github.com/openai/guided-diffusion>)

Recorded evaluation configuration: {"Total Images(M)":507,"Total Gflops":"3.76 x 10^11","Classifier-free Guidance":false}. This record does not assert independent reproduction.

### LDM-8-G

FID: 7.76. Result ID: 6991ad693a6c31484909d775.

[Reporting paper](<https://www.sota2.com/research/paper/68fa715b1e209131cea50d2a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c391e209131cea32e4a.md>)

[Implementation](<https://github.com/CompVis/latent-diffusion>)

Recorded evaluation configuration: {"Total Images(M)":307,"Total Gflops":"1.75 x 10^10","Classifier-free Guidance":true}. This record does not assert independent reproduction.

### DIFFUSSM-XL

FID: 9.07. Result ID: 6991ad693a6c31484909d772.

[Reporting paper](<https://www.sota2.com/research/paper/68fa715b1e209131cea50d2a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa715b1e209131cea50d2a.md>)

Recorded evaluation configuration: {"Total Images(M)":660,"Total Gflops":"1.85 x 10^11","Classifier-free Guidance":false}. This record does not assert independent reproduction.

### DiT-XL/2

FID: 9.62. Result ID: 6991ad693a6c31484909d771.

[Reporting paper](<https://www.sota2.com/research/paper/68fa715b1e209131cea50d2a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ea91e209131cea41328.md>)

Recorded evaluation configuration: {"Total Images(M)":1792,"Total Gflops":"2.13 x 10^11","Classifier-free Guidance":false}. This record does not assert independent reproduction.

### LDM-4

FID: 10.56. Result ID: 6991ad693a6c31484909d770.

[Reporting paper](<https://www.sota2.com/research/paper/68fa715b1e209131cea50d2a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c391e209131cea32e4a.md>)

[Implementation](<https://github.com/CompVis/latent-diffusion>)

Recorded evaluation configuration: {"Total Images(M)":213,"Total Gflops":"2.22 x 10^10","Classifier-free Guidance":false}. This record does not assert independent reproduction.

### ADM

FID: 10.94. Result ID: 6991ad693a6c31484909d76c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa715b1e209131cea50d2a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a0e1e209131cea25dba.md>)

[Implementation](<https://github.com/openai/guided-diffusion>)

Recorded evaluation configuration: {"Total Images(M)":507,"Total Gflops":"5.68 x 10^12","Classifier-free Guidance":false}. This record does not assert independent reproduction.

### LDM-8

FID: 15.51. Result ID: 6991ad693a6c31484909d76f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa715b1e209131cea50d2a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c391e209131cea32e4a.md>)

[Implementation](<https://github.com/CompVis/latent-diffusion>)

Recorded evaluation configuration: {"Total Images(M)":307,"Total Gflops":"1.75 x 10^10","Classifier-free Guidance":false}. This record does not assert independent reproduction.

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