# Text-to-Image Generation on DPG-Bench (DPG-Bench Metric) Source: [Text-to-Image Generation on DPG-Bench (DPG-Bench Metric)]() Collection index: [llms.txt]() Data updated: 2026-06-11T09:20:57.305Z Evaluation of Text-to-Image generation models on the DPG-Bench dataset, reporting performance using the specific DPG-Bench 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 | | Dataset (including any stated split) | DPG-Bench | | Metric | DPG-Bench Score | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 27 of 27 | | 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 - [DPG-Bench Score](): Higher is better; unit: PERCENT. ## Recorded results | Method | DPG-Bench Score | Result date | Report paper | | --- | --- | --- | --- | | Qwen-Image | 88.32 | Not recorded | 6a2a40ebd33dfb23e7839188 | | Seedream 3.0 | 88.27 | Not recorded | 6a2a40ebd33dfb23e7839188 | | Z-Image | 88.14 | Not recorded | 6a2a40ebd33dfb23e7839188 | | Lumina-Image 2.0 | 87.2 | Not recorded | 6a2a40ebd33dfb23e7839188 | | i1 | 86.73 | Not recorded | 6a2a40ebd33dfb23e7839188 | | Show-o2 | 86.14 | Not recorded | 6a02b424dbc836b87d8246e4 | | HiDream-I1-Full | 85.89 | Not recorded | 6a2a40ebd33dfb23e7839188 | | w/ RPO-Gemini 3.1 Pro-ARR | 85.76 | Not recorded | 6a02b424dbc836b87d8246e4 | | BAGEL | 85.44 | Not recorded | 6a2a40ebd33dfb23e7839188 | | w/ RPO-GPT-5-ARR | 85.41 | Not recorded | 6a02b424dbc836b87d8246e4 | | GPT Image 1 \[High\] | 85.15 | Not recorded | 6a2a40ebd33dfb23e7839188 | | BAGEL | 85.07 | Not recorded | 6a02b424dbc836b87d8246e4 | | w/ RPO-Qwen3vl-8B-ARR | 85.03 | Not recorded | 6a02b424dbc836b87d8246e4 | | Janus-Pro-7B | 84.19 | Not recorded | 6a2a40ebd33dfb23e7839188 | | SD3 Medium | 84.08 | Not recorded | 6a2a40ebd33dfb23e7839188 | | FLUX.1-Dev | 83.84 | Not recorded | 6a02b424dbc836b87d8246e4 | | FLUX.1 \[Dev\] | 83.84 | Not recorded | 6a2a40ebd33dfb23e7839188 | | OmniGen2 | 83.57 | Not recorded | 6a02b424dbc836b87d8246e4 | | DALLE-3 | 83.5 | Not recorded | 6a02b424dbc836b87d8246e4 | | BLIP3o-N-S | 81.98 | Not recorded | 6a2a40ebd33dfb23e7839188 | | BLIP3o-N-G-G | 81.93 | Not recorded | 6a2a40ebd33dfb23e7839188 | | DeCo | 81.4 | Not recorded | 6a2a40ebd33dfb23e7839188 | | PixNerd | 80.9 | Not recorded | 6a2a40ebd33dfb23e7839188 | | Emu3 | 80.6 | Not recorded | 6a02b424dbc836b87d8246e4 | | BLIP3o-N-G-T | 79.77 | Not recorded | 6a2a40ebd33dfb23e7839188 | | BLIP3o-4B | 79.73 | Not recorded | 6a2a40ebd33dfb23e7839188 | | JanusFlow | 79.68 | Not recorded | 6a02b424dbc836b87d8246e4 | ### Qwen-Image DPG-Bench Score: 88.32. Result ID: 6a2a7df3fbc373592c27e71a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"#params":"20B"}. This record does not assert independent reproduction. ### Seedream 3.0 DPG-Bench Score: 88.27. Result ID: 6a2a7df3fbc373592c27e712. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Z-Image DPG-Bench Score: 88.14. Result ID: 6a2a7df3fbc373592c27e719. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"#params":"6B"}. This record does not assert independent reproduction. ### Lumina-Image 2.0 DPG-Bench Score: 87.2. Result ID: 6a2a7df3fbc373592c27e718. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"#params":"3B"}. This record does not assert independent reproduction. ### i1 DPG-Bench Score: 86.73. Result ID: 6a2a7df3fbc373592c27e721. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"#params":"3B"}. This record does not assert independent reproduction. ### Show-o2 DPG-Bench Score: 86.14. Result ID: 6a02e39f47f34ce924e59f07. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### HiDream-I1-Full DPG-Bench Score: 85.89. Result ID: 6a2a7df3fbc373592c27e717. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"#params":"17B"}. This record does not assert independent reproduction. ### w/ RPO-Gemini 3.1 Pro-ARR DPG-Bench Score: 85.76. Result ID: 6a02e39f47f34ce924e59f0c. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Base Generator":"FLUX.1-Dev"}. This record does not assert independent reproduction. ### BAGEL DPG-Bench Score: 85.44. Result ID: 6a2a7df3fbc373592c27e716. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"#params":"14B"}. This record does not assert independent reproduction. ### w/ RPO-GPT-5-ARR DPG-Bench Score: 85.41. Result ID: 6a02e39f47f34ce924e59f0b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Base Generator":"FLUX.1-Dev"}. This record does not assert independent reproduction. ### GPT Image 1 \[High\] DPG-Bench Score: 85.15. Result ID: 6a2a7df3fbc373592c27e711. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### BAGEL DPG-Bench Score: 85.07. Result ID: 6a02e39f47f34ce924e59f09. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### w/ RPO-Qwen3vl-8B-ARR DPG-Bench Score: 85.03. Result ID: 6a02e39f47f34ce924e59f0a. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Base Generator":"FLUX.1-Dev"}. This record does not assert independent reproduction. ### Janus-Pro-7B DPG-Bench Score: 84.19. Result ID: 6a2a7df3fbc373592c27e715. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"#params":"7B"}. This record does not assert independent reproduction. ### SD3 Medium DPG-Bench Score: 84.08. Result ID: 6a2a7df3fbc373592c27e714. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"#params":"2B"}. This record does not assert independent reproduction. ### FLUX.1-Dev DPG-Bench Score: 83.84. Result ID: 6a02e39f47f34ce924e59f05. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### FLUX.1 \[Dev\] DPG-Bench Score: 83.84. Result ID: 6a2a7df3fbc373592c27e713. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"#params":"12B"}. This record does not assert independent reproduction. ### OmniGen2 DPG-Bench Score: 83.57. Result ID: 6a02e39f47f34ce924e59f08. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### DALLE-3 DPG-Bench Score: 83.5. Result ID: 6a02e39f47f34ce924e59f06. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### BLIP3o-N-S DPG-Bench Score: 81.98. Result ID: 6a2a7df3fbc373592c27e71e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"#params":"3B"}. This record does not assert independent reproduction. ### BLIP3o-N-G-G DPG-Bench Score: 81.93. Result ID: 6a2a7df3fbc373592c27e71f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"#params":"3B"}. This record does not assert independent reproduction. ### DeCo DPG-Bench Score: 81.4. Result ID: 6a2a7df3fbc373592c27e71d. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"#params":"1B"}. This record does not assert independent reproduction. ### PixNerd DPG-Bench Score: 80.9. Result ID: 6a2a7df3fbc373592c27e71c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"#params":"1B"}. This record does not assert independent reproduction. ### Emu3 DPG-Bench Score: 80.6. Result ID: 6a02e39f47f34ce924e59f03. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### BLIP3o-N-G-T DPG-Bench Score: 79.77. Result ID: 6a2a7df3fbc373592c27e720. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"#params":"3B"}. This record does not assert independent reproduction. ### BLIP3o-4B DPG-Bench Score: 79.73. Result ID: 6a2a7df3fbc373592c27e71b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"#params":"4B"}. This record does not assert independent reproduction. ### JanusFlow DPG-Bench Score: 79.68. Result ID: 6a02e39f47f34ce924e59f04. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.