# gpt-image

> **TL;DR.** **gpt-image** is OpenAI's natively multimodal image generation and editing API family — the same engine that powers ChatGPT's image tool, exposed as model IDs `gpt-image-1`, `gpt-image-1-mini`, and `gpt-image-1.

- **Category:** AI / Image Generation / APIs
- **Stage:** established
- **Age:** 419 days
- **Origin date:** 2025-04-23
- **First detected:** 2026-04-21
- **Canonical URL:** https://earlyterms.com/term/gpt-image
- **Sources:** 10 primary URLs

## Definition

**gpt-image** is OpenAI's natively multimodal image generation and editing API family — the same engine that powers ChatGPT's image tool, exposed as model IDs `gpt-image-1`, `gpt-image-1-mini`, and `gpt-image-1.5`. Unlike DALL-E 3's diffusion pipeline, the models render text, logos, and branded UI with high typographic fidelity and accept reference images for conversational edits.

The API [launched April 23, 2025](https://techcrunch.com/2025/04/23/openai-makes-its-upgraded-image-generator-available-to-developers/) with Adobe, Airtable, Canva, Figma, GoDaddy, Instacart, and Wix as day-one partners. OpenAI shipped [gpt-image-1-mini](https://simonwillison.net/2025/Oct/6/gpt-image-1-mini/) at DevDay on October 6 2025 at 80% lower cost, then [gpt-image-1.5](https://community.openai.com/t/gpt-image-1-5-rolling-out-in-the-api-and-chatgpt/1369443) on December 16 2025 with 4x faster generation and a 20% API price cut.

## Analogy

If DALL-E 3 was a painter who mangled street signs, gpt-image is a sign-painter who also paints.

## Why it's emerging now

LM Arena watchers caught three anonymous image models — `packingtape-alpha`, `maskingtape-alpha`, `gaffertape-alpha` — on April 4 2026, widely believed to be gpt-image-2 in grayscale test. ChatGPT Plus/Pro users began reporting A/B rollouts April 19. DALL-E 2 and 3 are scheduled for shutdown May 12, consolidating the entire OpenAI image stack under the `gpt-image` name.

## Related terms

- *parent:* DALL-E 3
- *child:* gpt-image-1
- *child:* gpt-image-1.5
- *child:* gpt-image-2
- *competitor:* Nano Banana
- *competitor:* Midjourney
- *competitor:* Seedream
- *competitor:* Stable Diffusion
- *alias:* ChatGPT Images
- *alias:* GPT-4o image generation
- *related:* C2PA metadata
- *related:* multimodal model

## Sources

1. [OpenAI — Introducing image generation in the API](https://openai.com/index/image-generation-api/)
2. [OpenAI — The new ChatGPT Images is here (gpt-image-1.5)](https://openai.com/index/new-chatgpt-images-is-here/)
3. [TechCrunch — OpenAI makes its upgraded image generator available to developers](https://techcrunch.com/2025/04/23/openai-makes-its-upgraded-image-generator-available-to-developers/)
4. [VentureBeat — OpenAI makes ChatGPT's image generation available as API](https://venturebeat.com/ai/openai-makes-chatgpts-image-generation-available-as-api)
5. [SiliconANGLE — OpenAI launches new GPT Image 1.5 model optimized for image editing](https://siliconangle.com/2025/12/16/openai-launches-new-gpt-image-1-5-model-optimized-image-editing/)
6. [Simon Willison — gpt-image-1-mini at DevDay 2025](https://simonwillison.net/2025/Oct/6/gpt-image-1-mini/)
7. [Wikipedia — GPT Image](https://en.wikipedia.org/wiki/GPT_Image)
8. [OpenAI Developer Community — GPT-Image-1.5 rolling out](https://community.openai.com/t/gpt-image-1-5-rolling-out-in-the-api-and-chatgpt/1369443)
9. [Microsoft Azure — Unveiling GPT-image-1 in Azure AI Foundry](https://azure.microsoft.com/en-us/blog/unveiling-gpt-image-1-rising-to-new-heights-with-image-generation-in-azure-ai-foundry/)
10. [gpt-image-1 API model page](https://platform.openai.com/docs/models/gpt-image-1)

---
_Generated by EarlyTerms · https://earlyterms.com/term/gpt-image_