# Unconditional Image Generation on ImageNet 64x64 (train)

Source: [Unconditional Image Generation on ImageNet 64x64 (train)](<https://www.sota2.com/research/sota/unconditional-image-generation-on-imagenet-64x64-train>)

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

Data updated: 2026-02-26T05:53:17.192Z

Evaluation of unconditional image generation models on the training split of ImageNet rendered at 64x64 resolution, typically measured by NLL (bits/dim).

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 | Unconditional Image Generation |
| Dataset (including any stated split) | ImageNet 64x64 (train) |
| Metric | NLL (bits/dim) |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 3 of 3 |
| 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

- [NLL (bits/dim)](<https://www.sota2.com/research/sota/unconditional-image-generation-on-imagenet-64x64-train.md?metric=NLL%20(bits%2Fdim)>): Lower is better; unit: SCALAR.

## Recorded results

| Method | NLL (bits/dim) | Result date | Report paper |
| --- | --- | --- | --- |
| Gated PixelCNN | 3.48 | Not recorded | 68fa629e1e209131ce9fa3c3 |
| PixelRNN | 3.57 | Not recorded | 68fa629e1e209131ce9fa3c3 |
| Conv Draw | 4.04 | Not recorded | 68fa629e1e209131ce9fa3c3 |

### Gated PixelCNN

NLL (bits/dim): 3.48. Result ID: 698bdc94ecace47ef5594748.

[Reporting paper](<https://www.sota2.com/research/paper/68fa629e1e209131ce9fa3c3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa629e1e209131ce9fa3c3.md>)

Recorded evaluation configuration: {"Resolution":"64x64","Batch size":128,"Layers":20,"Hidden units":384,"Filter size":"5x5"}. This record does not assert independent reproduction.

### PixelRNN

NLL (bits/dim): 3.57. Result ID: 698bdc94ecace47ef5594747.

[Reporting paper](<https://www.sota2.com/research/paper/68fa629e1e209131ce9fa3c3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa62b81e209131ce9fad7e.md>)

Recorded evaluation configuration: {"Resolution":"64x64"}. This record does not assert independent reproduction.

### Conv Draw

NLL (bits/dim): 4.04. Result ID: 698bdc94ecace47ef5594746.

[Reporting paper](<https://www.sota2.com/research/paper/68fa629e1e209131ce9fa3c3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa62b81e209131ce9fad7e.md>)

Recorded evaluation configuration: {"Resolution":"64x64"}. This record does not assert independent reproduction.

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