Official

openai / dall-e-2

The original classic DALLᐧE 2

  • Public
  • 61 runs
  • Paper
  • License
Iterate in playground
Run with an API
Playground API Examples README

Examples

View more

Pricing

Official model
Pricing for official models works differently from other models. Instead of being billed by time, you’re billed by input and output, making pricing more predictable.
Learn more

This model is priced by how many images are generated.

Check out our docs for more information about how per-image pricing works on Replicate.

Readme

DALL·E 2

DALL·E 2 is an AI system that creates realistic images and art from natural language descriptions. It can combine concepts, attributes, and styles to generate original content based on text prompts.

Capabilities

Creates original images based on detailed text descriptions. For example, typing “a photorealistic image of an astronaut riding a horse” will generate various interpretations of this concept.

Technical Details

  • Launched in 2022 as the successor to DALL·E (2021)
  • 4x greater resolution than the original DALL·E
  • Significantly improved photorealism and caption matching
  • Based on CLIP latents (Contrastive Language-Image Pre-training)

Safety Features

  • Content filtering to prevent generation of violent, hateful, or adult images
  • Restrictions on photorealistic generations of real individuals’ faces
  • Content policy prohibiting violent, adult, or political content generation
  • Automated and human monitoring systems

Limitations

  • Cannot generate all types of content due to safety restrictions
  • Subject to content policy restrictions
  • Other limitations documented in the risks and limitations documentation

Additional Resources