Statements (32)
| Predicate | Object |
|---|---|
| gptkbp:instanceOf |
gptkb:text-to-image_AI_model
|
| gptkbp:abilities |
inpainting
photorealistic image generation image-to-image generation outpainting high-fidelity image synthesis image variation text-guided image editing |
| gptkbp:architecture |
diffusion transformer
|
| gptkbp:developedBy |
gptkb:Google
|
| gptkbp:input |
text prompt
|
| gptkbp:language |
English
|
| gptkbp:notableFor |
efficient training
safety mitigations fine-grained text-image alignment large-scale dataset training responsible AI practices scalable model size state-of-the-art image quality |
| gptkbp:organization |
gptkb:Google_DeepMind
|
| gptkbp:output |
gptkb:illustrator
|
| gptkbp:predecessor |
gptkb:Imagen
|
| gptkbp:purpose |
text-to-image generation
|
| gptkbp:relatedTo |
gptkb:Midjourney
gptkb:Stable_Diffusion gptkb:DALL-E_3 |
| gptkbp:releaseDate |
2024
|
| gptkbp:researchPaper |
https://arxiv.org/abs/2405.20149
|
| gptkbp:uses |
diffusion model
|
| gptkbp:bfsParent |
gptkb:Google_Bard
|
| gptkbp:bfsLayer |
7
|
| https://www.w3.org/2000/01/rdf-schema#label |
Imagen 2
|