DALL·E needs a big UPDATE

I have been using DALL·E 3 within ChatGPT Plus for almost two years, and while GPT has improved significantly (now GPT-5), the DALL·E image generation model has not received any major updates since its release in October 2023.

As a paying subscriber, I expect the included image generation tool to evolve at the same pace as the text models. DALL·E 3 still struggles with:

– Preserving exact character designs from reference images.

– Maintaining consistent art styles (e.g., matching a specific splash art style).

– Accurately reproducing colors and fine details without distortion.

Competing services (MidJourney, Leonardo AI, Stable Diffusion) have made frequent and substantial improvements in this period, and DALL·E is falling behind.

I kindly request that OpenAI prioritize a major update to DALL·E — whether it’s an improved DALL·E 3 or the release of DALL·E 4 — with better reference fidelity, style matching, and color accuracy. This would ensure the image generation tool matches the quality and progress of GPT-5.

Thank you for your attention to this matter.

Hi,

https://openai.com/index/introducing-4o-image-generation/.

There has been a recent update to images.

Check out some images contributed by forum members here.

The Official 4o and Dall-E image Megathread.