# Text-to-Image Generation on GenEval (GenEval Score and R Metric)

Source: [Text-to-Image Generation on GenEval (GenEval Score and R Metric)](<https://www.sota2.com/research/sota/text-to-image-generation-on-geneval-geneval-score-and-r-metric>)

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

Data updated: 2026-02-26T05:45:55.057Z

Evaluation of Text-to-Image generation models on the GenEval dataset, reporting the primary GenEval score and the R 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 | GenEval Score |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 16 of 16 |
| 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

- [GenEval Score](<https://www.sota2.com/research/sota/text-to-image-generation-on-geneval-geneval-score-and-r-metric.md?metric=GenEval%20Score>): Higher is better; unit: PERCENT.

- [R Metric (%)](<https://www.sota2.com/research/sota/text-to-image-generation-on-geneval-geneval-score-and-r-metric.md?metric=R%20Metric%20(%25)>): Lower is better; unit: PERCENT.

## Recorded results

| Method | GenEval Score | Result date | Report paper |
| --- | --- | --- | --- |
| Base model | 87 | Not recorded | 691ff240d6b0dd334292e89e |
| PPCL(14B) | 84.7 | Not recorded | 691ff240d6b0dd334292e89e |
| PPCL(Fine-tuning) | 82.8 | Not recorded | 691ff240d6b0dd334292e89e |
| PPCL(12B) | 80.1 | Not recorded | 691ff240d6b0dd334292e89e |
| PPCL(10B) | 78.4 | Not recorded | 691ff240d6b0dd334292e89e |
| HierarchicalPrune | 76.6 | Not recorded | 691ff240d6b0dd334292e89e |
| TinyFusion | 73.9 | Not recorded | 691ff240d6b0dd334292e89e |
| Base model | 66.5 | Not recorded | 691ff240d6b0dd334292e89e |
| FLUX.1 Lite | 62.3 | Not recorded | 691ff240d6b0dd334292e89e |
| Base model | 62.3 | Not recorded | 691ff240d6b0dd334292e89e |
| PPCL(8B) | 60.5 | Not recorded | 691ff240d6b0dd334292e89e |
| Chroma1-HD | 59.3 | Not recorded | 691ff240d6b0dd334292e89e |
| PPCL(6.5B) | 59.3 | Not recorded | 691ff240d6b0dd334292e89e |
| TinyFusion | 51.1 | Not recorded | 691ff240d6b0dd334292e89e |
| HierarchicalPrune | 50.3 | Not recorded | 691ff240d6b0dd334292e89e |
| Dense2MoE | 40.3 | Not recorded | 691ff240d6b0dd334292e89e |

### Base model

GenEval Score: 87. Result ID: 699ea3da5698665ec46ed1cb.

[Reporting paper](<https://www.sota2.com/research/paper/691ff240d6b0dd334292e89e.md>)

Recorded evaluation configuration: {"Model Context":"Qwen-Image","P.(B)↓":20,"M.(%)↓":100,"L.(ms)↓":2625}. This record does not assert independent reproduction.

### PPCL(14B)

GenEval Score: 84.7. Result ID: 699ea3da5698665ec46ed1ce.

[Reporting paper](<https://www.sota2.com/research/paper/691ff240d6b0dd334292e89e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/691ff240d6b0dd334292e89e.md>)

Recorded evaluation configuration: {"Model Context":"Qwen-Image","P.(B)↓":14,"M.(%)↓":79.4,"L.(ms)↓":1792}. This record does not assert independent reproduction.

### PPCL(Fine-tuning)

GenEval Score: 82.8. Result ID: 699ea3da5698665ec46ed1d1.

[Reporting paper](<https://www.sota2.com/research/paper/691ff240d6b0dd334292e89e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/691ff240d6b0dd334292e89e.md>)

Recorded evaluation configuration: {"Model Context":"Qwen-Image","P.(B)↓":10,"M.(%)↓":66.9,"L.(ms)↓":1462}. This record does not assert independent reproduction.

### PPCL(12B)

GenEval Score: 80.1. Result ID: 699ea3da5698665ec46ed1cf.

[Reporting paper](<https://www.sota2.com/research/paper/691ff240d6b0dd334292e89e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/691ff240d6b0dd334292e89e.md>)

Recorded evaluation configuration: {"Model Context":"Qwen-Image","P.(B)↓":12,"M.(%)↓":71.4,"L.(ms)↓":1514}. This record does not assert independent reproduction.

### PPCL(10B)

GenEval Score: 78.4. Result ID: 699ea3da5698665ec46ed1d0.

[Reporting paper](<https://www.sota2.com/research/paper/691ff240d6b0dd334292e89e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/691ff240d6b0dd334292e89e.md>)

Recorded evaluation configuration: {"Model Context":"Qwen-Image","P.(B)↓":10,"M.(%)↓":66.9,"L.(ms)↓":1462}. This record does not assert independent reproduction.

### HierarchicalPrune

GenEval Score: 76.6. Result ID: 699ea3da5698665ec46ed1cd.

[Reporting paper](<https://www.sota2.com/research/paper/691ff240d6b0dd334292e89e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a731e209131cea8352b.md>)

Recorded evaluation configuration: {"Model Context":"Qwen-Image","P.(B)↓":14,"M.(%)↓":79.4,"L.(ms)↓":1786}. This record does not assert independent reproduction.

### TinyFusion

GenEval Score: 73.9. Result ID: 699ea3da5698665ec46ed1cc.

[Reporting paper](<https://www.sota2.com/research/paper/691ff240d6b0dd334292e89e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76061e209131cea6afea.md>)

[Implementation](<https://github.com/VainF/TinyFusion>)

Recorded evaluation configuration: {"Model Context":"Qwen-Image","P.(B)↓":14,"M.(%)↓":79.4,"L.(ms)↓":1789}. This record does not assert independent reproduction.

### Base model

GenEval Score: 66.5. Result ID: 699ea3da5698665ec46ed1c2.

[Reporting paper](<https://www.sota2.com/research/paper/691ff240d6b0dd334292e89e.md>)

Recorded evaluation configuration: {"Model Context":"FLUX.1-dev","P.(B)↓":12,"M.(%)↓":100,"L.(ms)↓":715}. This record does not assert independent reproduction.

### FLUX.1 Lite

GenEval Score: 62.3. Result ID: 699ea3da5698665ec46ed1c3.

[Reporting paper](<https://www.sota2.com/research/paper/691ff240d6b0dd334292e89e.md>)

Recorded evaluation configuration: {"Model Context":"FLUX.1-dev","P.(B)↓":8,"M.(%)↓":78.8,"L.(ms)↓":572}. This record does not assert independent reproduction.

### Base model

GenEval Score: 62.3. Result ID: 699ea3da5698665ec46ed1c9.

[Reporting paper](<https://www.sota2.com/research/paper/691ff240d6b0dd334292e89e.md>)

Recorded evaluation configuration: {"Model Context":"FLUX.1 Lite","P.(B)↓":8,"M.(%)↓":78.8,"L.(ms)↓":572}. This record does not assert independent reproduction.

### PPCL(8B)

GenEval Score: 60.5. Result ID: 699ea3da5698665ec46ed1c8.

[Reporting paper](<https://www.sota2.com/research/paper/691ff240d6b0dd334292e89e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/691ff240d6b0dd334292e89e.md>)

Recorded evaluation configuration: {"Model Context":"FLUX.1-dev","P.(B)↓":8,"M.(%)↓":74.4,"L.(ms)↓":535}. This record does not assert independent reproduction.

### Chroma1-HD

GenEval Score: 59.3. Result ID: 699ea3da5698665ec46ed1c4.

[Reporting paper](<https://www.sota2.com/research/paper/691ff240d6b0dd334292e89e.md>)

Recorded evaluation configuration: {"Model Context":"FLUX.1-dev","P.(B)↓":8.9,"M.(%)↓":82.5,"L.(ms)↓":1714}. This record does not assert independent reproduction.

### PPCL(6.5B)

GenEval Score: 59.3. Result ID: 699ea3da5698665ec46ed1ca.

[Reporting paper](<https://www.sota2.com/research/paper/691ff240d6b0dd334292e89e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/691ff240d6b0dd334292e89e.md>)

Recorded evaluation configuration: {"Model Context":"FLUX.1 Lite","P.(B)↓":6.5,"M.(%)↓":69.2,"L.(ms)↓":428}. This record does not assert independent reproduction.

### TinyFusion

GenEval Score: 51.1. Result ID: 699ea3da5698665ec46ed1c6.

[Reporting paper](<https://www.sota2.com/research/paper/691ff240d6b0dd334292e89e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76061e209131cea6afea.md>)

[Implementation](<https://github.com/VainF/TinyFusion>)

Recorded evaluation configuration: {"Model Context":"FLUX.1-dev","P.(B)↓":8,"M.(%)↓":74.4,"L.(ms)↓":534}. This record does not assert independent reproduction.

### HierarchicalPrune

GenEval Score: 50.3. Result ID: 699ea3da5698665ec46ed1c7.

[Reporting paper](<https://www.sota2.com/research/paper/691ff240d6b0dd334292e89e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a731e209131cea8352b.md>)

Recorded evaluation configuration: {"Model Context":"FLUX.1-dev","P.(B)↓":8,"M.(%)↓":74.4,"L.(ms)↓":543}. This record does not assert independent reproduction.

### Dense2MoE

GenEval Score: 40.3. Result ID: 699ea3da5698665ec46ed1c5.

[Reporting paper](<https://www.sota2.com/research/paper/691ff240d6b0dd334292e89e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bff1e209131cea8bc4a.md>)

Recorded evaluation configuration: {"Model Context":"FLUX.1-dev","P.(B)↓":12,"M.(%)↓":100,"L.(ms)↓":312}. This record does not assert independent reproduction.

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