OpenAI DALL·E API public beta announcement

TL;DR

OpenAI announced that the DALL·E image generation API is now available in public beta, enabling developers to embed high‑quality, text‑to‑image generation with built‑in safety filters into their applications.

Public beta launch

The DALL·E API launched today in public beta, allowing developers to call the service via a simple HTTP endpoint. The announcement includes a sample curl request that generates a 1024×1024 image from a natural‑language prompt. The API is intended to be usable within minutes of obtaining an OpenAI API key.

curl https://api.openai.com/v1/images/generations \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $OPENAI_API_KEY" \
  -d '{
    "prompt": "a photo of a happy corgi puppy sitting and facing forward, studio light, longshot",
    "n":1,
    "size":"1024x1024"
   }'

State‑of‑the‑art image generation

DALL·E’s core capability is to create and edit original images from textual descriptions, covering styles from artistic illustrations to photorealistic scenes. The model is praised for its ability to follow natural language prompts accurately. OpenAI notes that future API updates will focus on improving image quality, latency, scalability, and overall usability.

Built‑in moderation

The API includes automatic moderation filters that block hate symbols, gore, and other disallowed content. OpenAI emphasizes that these safeguards are derived from trust‑and‑safety lessons learned while serving more than three million DALL·E users, who collectively generate over four million images per day. The company commits to continuing safety improvements so developers can focus on product development.

Early adopters and use cases

OpenAI highlighted three early customers that have integrated DALL·E into their products:

  • Microsoft Designer – A graphic‑design tool for social media posts, invitations, and digital postcards. Microsoft also embeds DALL·E in Bing and Microsoft Edge via an "Image Creator" feature that generates images when search results are insufficient.
  • CALA – Described as the world’s first fashion and lifestyle operating system, CALA uses DALL·E to let users generate design concepts from text or reference images, streamlining the workflow from ideation to e‑commerce.
  • Mixtiles – A photo‑printing service that leverages the DALL·E API to produce custom artwork for wall displays, guiding users through creative prompts that capture personal memories or imagined scenes.

These examples illustrate how DALL·E can power both creative design tools and consumer‑facing experiences.

Integration with OpenAI’s broader platform

The DALL·E API is part of OpenAI’s larger API platform, which also provides access to language models and other multimodal capabilities. Developers can combine DALL·E with text‑generation models to build richer, multimodal applications.

What to expect next

OpenAI states that the API will continue to evolve, with planned enhancements to image fidelity, generation speed, and safety mechanisms. The public beta invites developers to experiment, provide feedback, and contribute to the roadmap for future releases.

Sources