# DALL-E (Agent)

> Source: https://aiwiki.ai/wiki/dall-e_agent
> Updated: 2026-06-07
> Categories: ChatGPT, Image Generation, OpenAI
> From AI Wiki (https://aiwiki.ai), a free encyclopedia of artificial intelligence. Quote with attribution.

DALL-E (Agent) refers to the official DALL·E [GPT](/wiki/gpts) published by [OpenAI](/wiki/openai) inside [ChatGPT](/wiki/chatgpt), a first-party agent variant of ChatGPT that exposes DALL·E 3 image generation through a dedicated conversational interface. The agent is reachable at the URL `chatgpt.com/g/g-2fkFE8rbu-dall-e` and is operated by OpenAI rather than a third-party developer.[^1] It is distinct from DALL·E itself, the underlying family of text-to-image diffusion and transformer models, which is covered at [dall e](/wiki/dall_e). The DALL·E GPT became a separate, persistent endpoint after OpenAI replaced DALL·E 3 as ChatGPT's default image generator with GPT Image (the GPT-4o-native image stack) in March 2025, leaving the dedicated GPT as the primary way for ChatGPT users to reach the older DALL·E 3 model from the consumer product.[^2]

## What this slug refers to

This page disambiguates the agent or wrapper from the model. The DALL·E model (text-to-image generation, originally a 12-billion-parameter [GPT-3](/wiki/gpt_4o)-derived system, later succeeded by DALL·E 2 and DALL·E 3) is documented at [dall e](/wiki/dall_e) and the third-generation variant at [dall e 3](/wiki/dall_e_3). The DALL·E GPT, by contrast, is a productized chat agent that lives in the [ChatGPT Store](/wiki/chatgpt_store) (see also [gpt store](/wiki/gpt_store)) and that uses the conversational tool-use loop introduced with ChatGPT's image-generation tool in late 2023 to route prompts to the underlying image model.[^3][^4]

The distinction matters because the agent and the model have separate launch dates, separate update cadences, and separate user experiences. The model launched in three generations between January 2021 (DALL·E 1) and late 2023 (DALL·E 3).[^5] The agent, in the sense of a first-class GPT object selectable from the ChatGPT GPT picker, only became prominent after the [GPT Store](/wiki/gpt_store) opened on January 10, 2024, and grew in importance after March 2025, when DALL·E 3 was no longer the default image generator inside the main ChatGPT chat.[^2][^6]

## DALL-E as integrated agent / GPT in ChatGPT

OpenAI announced DALL·E 3 on September 20, 2023, and made it generally available inside [ChatGPT](/wiki/chatgpt) Plus and ChatGPT Enterprise in early October 2023.[^7][^8] In that initial integration, DALL·E 3 was not a separate GPT; it was a tool callable from the default ChatGPT conversation, with the language model acting as a prompt engineer that turned the user's request into a longer, more descriptive caption before sending it to the image model.[^4] OpenAI described this as part of an "ask-as-you-would-a-person" workflow in which the user does not need to write image-style prompts directly.[^4]

When OpenAI launched the [GPT Store](/wiki/gpt_store) on January 10, 2024, [GPTs](/wiki/gpts) (called custom GPTs at the time, see [custom gpt](/wiki/custom_gpt) and [custom gpts](/wiki/custom_gpts)) became browseable, shareable, and rankable agents.[^6] OpenAI itself maintained a small number of first-party GPTs, of which the DALL·E GPT is one. The DALL·E GPT carries a builder attribution of "By ChatGPT" or "By OpenAI" and uses the same image-generation tool that the broader ChatGPT product calls, but it is configured with a system prompt that focuses the conversation on image creation and revision rather than open-ended chat.[^1][^9]

The DALL·E GPT became materially more visible in March 2025, when OpenAI rolled out GPT Image (the [GPT-4o](/wiki/gpt_4o) native image generator) as the default image-generation path inside ChatGPT for Plus, Pro, Team, and Free users.[^2] DALL·E 3 was no longer reachable by default from ordinary chats, but it remained available to ChatGPT users who explicitly opened the DALL·E GPT, and it remained available through the [OpenAI API](/wiki/openai_api).[^2] In effect, the DALL·E GPT became the canonical consumer surface for the DALL·E 3 model after that switchover, while [AI image generation](/wiki/ai_image_generation) in the default ChatGPT experience was handled by GPT Image.

## Features

The DALL·E GPT is a thin agent over DALL·E 3. Its capabilities are therefore the capabilities of the underlying model, plus the prompting and tool-use behaviour OpenAI has wired into its system prompt. Users open the GPT from the ChatGPT app, the web sidebar, or by direct URL, then describe an image in natural language. The agent rewrites the request into a longer prompt of roughly 100 words, calls the image tool, and returns one or two generated images per turn.[^9] DALL·E 3 itself was trained to follow long prompts closely, to render text inside images (a known weakness of DALL·E 2), and to produce coherent compositions with multiple subjects.[^7]

Reported user-facing features include:

- Prompt expansion. The agent automatically converts short user descriptions into detailed image prompts, removing the need for hand-written prompt engineering.[^4]
- Iterative regeneration. Users can ask for variants, style changes, recompositions, or specific element edits, and the agent issues fresh image calls; ChatGPT keeps prior images in the conversation for reference.[^10]
- Style and aspect-ratio control. The agent accepts requests for square (1024 x 1024), portrait (1024 x 1792), or landscape (1792 x 1024) images, matching DALL·E 3's API sizes, and accepts style hints such as "photorealistic," "watercolor," or "vector illustration."[^11]
- Rate limits aligned with ChatGPT Plus. The DALL·E GPT inherits ChatGPT's per-user image quotas; reporting and OpenAI community documentation place the Plus tier at roughly 50 image generations per three-hour rolling window.[^12]
- Built-in policy guardrails. The agent's system prompt blocks images of named living public figures and copyrighted characters, restricts artist-style emulation for living artists, and applies content-safety classifiers consistent with the DALL·E 3 system card.[^9][^13]
- Conversational handoff. Because the DALL·E GPT runs on top of ChatGPT, the same conversation can hold both image generations and ordinary text, and the user can ask the agent to caption, describe, or critique the images it has just produced.[^4]

The DALL·E GPT does not expose finer-grained model parameters (sampler choice, guidance scale, seed) the way the [OpenAI API](/wiki/openai_api) does; those controls remain in the API-level `images/generations` endpoint.[^11]

## See also

- [dall e](/wiki/dall_e)
- [dall-e](/wiki/dall-e)
- [dall e 3](/wiki/dall_e_3)
- [chatgpt](/wiki/chatgpt)
- [gpts](/wiki/gpts)
- [chatgpt store](/wiki/chatgpt_store)
- [gpt store](/wiki/gpt_store)
- [custom gpt](/wiki/custom_gpt)
- [custom gpts](/wiki/custom_gpts)
- [dall e custom gpts](/wiki/dall_e_custom_gpts)
- [gpt 4o](/wiki/gpt_4o)
- [openai](/wiki/openai)
- [ai image generation](/wiki/ai_image_generation)
- [ai agents](/wiki/ai_agents)

## References

[^1]: OpenAI, "DALL·E GPT", ChatGPT GPT Store, 2024. https://chatgpt.com/g/g-2fkFE8rbu-dall-e. Accessed 2026-05-26.
[^2]: OpenAI, "Introducing 4o Image Generation", OpenAI blog, 2025-03-25. https://openai.com/index/introducing-4o-image-generation/. Accessed 2026-05-26.
[^3]: Wikipedia contributors, "DALL-E", Wikipedia, 2026. https://en.wikipedia.org/wiki/DALL-E. Accessed 2026-05-26.
[^4]: OpenAI, "DALL·E 3 is now available in ChatGPT Plus and Enterprise", OpenAI blog, 2023-10-19. https://openai.com/index/dall-e-3-is-now-available-in-chatgpt-plus-and-enterprise/. Accessed 2026-05-26.
[^5]: OpenAI, "DALL·E: Creating images from text", OpenAI blog, 2021-01-05. https://openai.com/index/dall-e/. Accessed 2026-05-26.
[^6]: Kyle Wiggers, "OpenAI launches a store for custom AI-powered chatbots", TechCrunch, 2024-01-10. https://techcrunch.com/2024/01/10/openai-launches-a-store-for-custom-ai-powered-chatbots/. Accessed 2026-05-26.
[^7]: OpenAI, "DALL·E 3", OpenAI blog, 2023-09-20. https://openai.com/index/dall-e-3/. Accessed 2026-05-26.
[^8]: Kyle Wiggers, "OpenAI unveils DALL-E 3, allows artists to opt out of training", TechCrunch, 2023-09-20. https://techcrunch.com/2023/09/20/openai-unveils-dall-e-3-allows-artists-to-opt-out-of-training/. Accessed 2026-05-26.
[^9]: Dustin Miller, "ChatGPT-AutoExpert: dalle.md (mirrored DALL·E GPT system prompt)", GitHub, 2024. https://github.com/spdustin/ChatGPT-AutoExpert/blob/main/_system-prompts/gpts/dalle.md. Accessed 2026-05-26.
[^10]: Maria Diaz, "OpenAI rolls out new features to ChatGPT's DALL-E image generator", SiliconANGLE, 2024-04-03. https://siliconangle.com/2024/04/03/openai-rolls-new-features-chatgpts-dall-e-image-generator/. Accessed 2026-05-26.
[^11]: OpenAI, "Image generation", OpenAI API documentation, 2025. https://platform.openai.com/docs/guides/image-generation. Accessed 2026-05-26.
[^12]: OpenAI Developer Community, "Dall-e3 daily limit 20 images? Is this new?", OpenAI forum, 2024. https://community.openai.com/t/dall-e3-daily-limit-20-images-is-this-new/457734. Accessed 2026-05-26.
[^13]: OpenAI, "DALL·E 3 system card", OpenAI blog, 2023-10-03. https://openai.com/index/dall-e-3-system-card/. Accessed 2026-05-26.