# Text-to-Image Prompt Following on DenseAlignBench

Source: [Text-to-Image Prompt Following on DenseAlignBench](<https://www.sota2.com/research/sota/text-to-image-prompt-following-on-densealignbench>)

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

Data updated: 2026-04-15T09:32:44.362Z

Evaluation of Text-to-Image Prompt Following capabilities using the DenseAlignBench dataset, measured via pairwise comparison metrics such as Win Rate and Net Advantage.

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 | Text-to-Image Prompt Following |
| Dataset (including any stated split) | DenseAlignBench |
| Metric | Win Rate |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 2 of 2 |
| 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

- [Win Rate](<https://www.sota2.com/research/sota/text-to-image-prompt-following-on-densealignbench.md?metric=Win%20Rate>): Higher is better; unit: PERCENT.

- [Baseline Win Rate](<https://www.sota2.com/research/sota/text-to-image-prompt-following-on-densealignbench.md?metric=Baseline%20Win%20Rate>): Higher is better; unit: PERCENT.

- [Tie Rate](<https://www.sota2.com/research/sota/text-to-image-prompt-following-on-densealignbench.md?metric=Tie%20Rate>): Higher is better; unit: PERCENT.

- [Net Advantage](<https://www.sota2.com/research/sota/text-to-image-prompt-following-on-densealignbench.md?metric=Net%20Advantage>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Win Rate | Result date | Report paper |
| --- | --- | --- | --- |
| Z-Image + PromptEcho | 61.5 | Not recorded | 69df1bb97d0172140ec7baad |
| QwenImage-2512 + PromptEcho | 53.3 | Not recorded | 69df1bb97d0172140ec7baad |

### Z-Image + PromptEcho

Win Rate: 61.5. Result ID: 69df5afd47f34ce924c86e1e.

[Reporting paper](<https://www.sota2.com/research/paper/69df1bb97d0172140ec7baad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69df1bb97d0172140ec7baad.md>)

Recorded evaluation configuration: {"Base Model":"Z-Image","Evaluator":"Gemini-3-flash-preview","Resolution":"1024×1024","CFG":"4.0","Inference steps":"50"}. This record does not assert independent reproduction.

### QwenImage-2512 + PromptEcho

Win Rate: 53.3. Result ID: 69df5afd47f34ce924c86e1f.

[Reporting paper](<https://www.sota2.com/research/paper/69df1bb97d0172140ec7baad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69df1bb97d0172140ec7baad.md>)

Recorded evaluation configuration: {"Base Model":"QwenImage-2512","Evaluator":"Gemini-3-flash-preview","Resolution":"1024×1024","CFG":"4.0","Inference steps":"50"}. This record does not assert independent reproduction.

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