DALL·E
E17411
DALL·E is an AI model developed by OpenAI that generates images from natural language descriptions, enabling text-to-image synthesis.
Aliases (3)
Statements (49)
| Predicate | Object |
|---|---|
| instanceOf |
artificial intelligence model
→
text-to-image generative model → |
| announcedBy |
OpenAI
→
|
| announcedIn |
January 2021
→
|
| application |
concept art
→
creative design → illustration → rapid visual prototyping → |
| basedOn |
GPT architecture
→
transformer architecture → |
| capability |
combine disparate concepts in a single image
→
compose multiple objects with spatial relationships → generate novel images from textual prompts → modify attributes of objects described in text → render objects with different artistic styles → |
| computingPlatform |
cloud computing infrastructure
→
|
| deploymentForm |
research model
→
|
| describedIn |
OpenAI blog post "DALL·E: Creating Images from Text"
→
|
| developer |
OpenAI
→
|
| field |
computer vision
→
multimodal machine learning → natural language processing → |
| genre |
generative AI
→
|
| hasVersion |
DALL·E 2
→
DALL·E 3 → |
| inspiredBy |
GPT-3
→
|
| language |
English
→
|
| license |
proprietary
→
|
| mediaType |
software
→
|
| nameInspiredBy |
Salvador Dalí
→
WALL·E → |
| notableFor |
fine-grained control via text prompts
→
highly creative image generations → |
| organization |
OpenAI
→
|
| producesOutput |
digital images
→
|
| publisher |
OpenAI
→
|
| relatedTo |
CLIP
→
GPT-3 → |
| safetyConsideration |
bias mitigation
→
content filtering → |
| successor |
DALL·E 2
→
DALL·E 3 → |
| supportsInput |
text prompts
→
|
| task |
image generation from natural language descriptions
→
text-to-image synthesis → |
| trainingDataType |
image-text pairs
→
|
| uses |
autoregressive modeling
→
discrete image tokens → neural networks → |
Referenced by (9)
| Subject (surface form when different) | Predicate |
|---|---|
|
OpenAI
→
OpenAI ("DALL·E 2") → OpenAI ("DALL·E 3") → |
developed |
|
DALL·E
("DALL·E 2")
→
DALL·E ("DALL·E 3") → |
hasVersion |
|
DALL·E
("DALL·E 2")
→
DALL·E ("DALL·E 3") → |
successor |
|
DALL·E
("OpenAI blog post "DALL·E: Creating Images from Text"")
→
|
describedIn |
|
OpenAI
→
|
knownFor |