# Text-to-Image Generation on GenEval (VQAScore)

Source: [Text-to-Image Generation on GenEval (VQAScore)](<https://www.sota2.com/research/sota/text-to-image-generation-on-geneval-vqascore>)

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

Data updated: 2026-03-04T18:29:12.577Z

Evaluation of Text-to-Image generation models on the GenEval dataset, focusing on the VQAScore metric.

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 Generation |
| Dataset (including any stated split) | GenEval |
| Metric | VQAScore |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 14 of 14 |
| 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

- [VQAScore](<https://www.sota2.com/research/sota/text-to-image-generation-on-geneval-vqascore.md?metric=VQAScore>): Higher is better; unit: PERCENT.

## Recorded results

| Method | VQAScore | Result date | Report paper |
| --- | --- | --- | --- |
| P-GRAFT(0.25N) | 80.96 | Not recorded | 68fa7bcf1e209131cea8a9f1 |
| P-GRAFT(0.5N) | 80.6 | Not recorded | 68fa7bcf1e209131cea8a9f1 |
| P-GRAFT(0.25N) | 80.29 | Not recorded | 68fa7bcf1e209131cea8a9f1 |
| GRAFT | 79.85 | Not recorded | 68fa7bcf1e209131cea8a9f1 |
| GRAFT | 79.57 | Not recorded | 68fa7bcf1e209131cea8a9f1 |
| P-GRAFT(0.75N) | 79.44 | Not recorded | 68fa7bcf1e209131cea8a9f1 |
| P-GRAFT(0.5N) | 79.25 | Not recorded | 68fa7bcf1e209131cea8a9f1 |
| P-GRAFT(0.75N) | 78.5 | Not recorded | 68fa7bcf1e209131cea8a9f1 |
| SDXL-Base | 73.9 | Not recorded | 68fa7bcf1e209131cea8a9f1 |
| SDXL-Base | 73.9 | Not recorded | 68fa7bcf1e209131cea8a9f1 |
| SD v2 | 73.49 | Not recorded | 68fa7bcf1e209131cea8a9f1 |
| SD v2 | 73.49 | Not recorded | 68fa7bcf1e209131cea8a9f1 |
| DDPO | 72.13 | Not recorded | 68fa7bcf1e209131cea8a9f1 |
| DDPO | 69.6 | Not recorded | 68fa7bcf1e209131cea8a9f1 |

### P-GRAFT(0.25N)

VQAScore: 80.96. Result ID: 69a879e26dbb7075b9509e2d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bcf1e209131cea8a9f1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bcf1e209131cea8a9f1.md>)

Recorded evaluation configuration: {"Fine-tuning dataset":"GenAI-Bench"}. This record does not assert independent reproduction.

### P-GRAFT(0.5N)

VQAScore: 80.6. Result ID: 69a879e26dbb7075b9509e2c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bcf1e209131cea8a9f1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bcf1e209131cea8a9f1.md>)

Recorded evaluation configuration: {"Fine-tuning dataset":"GenAI-Bench"}. This record does not assert independent reproduction.

### P-GRAFT(0.25N)

VQAScore: 80.29. Result ID: 69a879e26dbb7075b9509e34.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bcf1e209131cea8a9f1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bcf1e209131cea8a9f1.md>)

Recorded evaluation configuration: {"Fine-tuning dataset":"T2ICompBench++ - Train"}. This record does not assert independent reproduction.

### GRAFT

VQAScore: 79.85. Result ID: 69a879e26dbb7075b9509e2a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bcf1e209131cea8a9f1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bcf1e209131cea8a9f1.md>)

Recorded evaluation configuration: {"Fine-tuning dataset":"GenAI-Bench"}. This record does not assert independent reproduction.

### GRAFT

VQAScore: 79.57. Result ID: 69a879e26dbb7075b9509e31.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bcf1e209131cea8a9f1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bcf1e209131cea8a9f1.md>)

Recorded evaluation configuration: {"Fine-tuning dataset":"T2ICompBench++ - Train"}. This record does not assert independent reproduction.

### P-GRAFT(0.75N)

VQAScore: 79.44. Result ID: 69a879e26dbb7075b9509e2b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bcf1e209131cea8a9f1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bcf1e209131cea8a9f1.md>)

Recorded evaluation configuration: {"Fine-tuning dataset":"GenAI-Bench"}. This record does not assert independent reproduction.

### P-GRAFT(0.5N)

VQAScore: 79.25. Result ID: 69a879e26dbb7075b9509e33.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bcf1e209131cea8a9f1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bcf1e209131cea8a9f1.md>)

Recorded evaluation configuration: {"Fine-tuning dataset":"T2ICompBench++ - Train"}. This record does not assert independent reproduction.

### P-GRAFT(0.75N)

VQAScore: 78.5. Result ID: 69a879e26dbb7075b9509e32.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bcf1e209131cea8a9f1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bcf1e209131cea8a9f1.md>)

Recorded evaluation configuration: {"Fine-tuning dataset":"T2ICompBench++ - Train"}. This record does not assert independent reproduction.

### SDXL-Base

VQAScore: 73.9. Result ID: 69a879e26dbb7075b9509e28.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bcf1e209131cea8a9f1.md>)

Recorded evaluation configuration: {"Fine-tuning dataset":"GenAI-Bench"}. This record does not assert independent reproduction.

### SDXL-Base

VQAScore: 73.9. Result ID: 69a879e26dbb7075b9509e2f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bcf1e209131cea8a9f1.md>)

Recorded evaluation configuration: {"Fine-tuning dataset":"T2ICompBench++ - Train"}. This record does not assert independent reproduction.

### SD v2

VQAScore: 73.49. Result ID: 69a879e26dbb7075b9509e27.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bcf1e209131cea8a9f1.md>)

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

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

Recorded evaluation configuration: {"Fine-tuning dataset":"GenAI-Bench"}. This record does not assert independent reproduction.

### SD v2

VQAScore: 73.49. Result ID: 69a879e26dbb7075b9509e2e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bcf1e209131cea8a9f1.md>)

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

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

Recorded evaluation configuration: {"Fine-tuning dataset":"T2ICompBench++ - Train"}. This record does not assert independent reproduction.

### DDPO

VQAScore: 72.13. Result ID: 69a879e26dbb7075b9509e29.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bcf1e209131cea8a9f1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71bd1e209131cea52dc8.md>)

[Implementation](<http://rl-diffusion.github.io/>)

Recorded evaluation configuration: {"Fine-tuning dataset":"GenAI-Bench"}. This record does not assert independent reproduction.

### DDPO

VQAScore: 69.6. Result ID: 69a879e26dbb7075b9509e30.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bcf1e209131cea8a9f1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71bd1e209131cea52dc8.md>)

[Implementation](<http://rl-diffusion.github.io/>)

Recorded evaluation configuration: {"Fine-tuning dataset":"T2ICompBench++ - Train"}. This record does not assert independent reproduction.

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