# ChatGPT 4o Image

> ChatGPT 4o Image is available on Artifio.ai — a unified AI workspace with a single pay-as-you-go wallet.

- Canonical page: https://artifio.ai/models/4o-image-api
- Run it: https://artifio.ai/create/4o-image-api
- Modality: image editing
- Model brand: openai
- Pricing: from about $0.030 per generation (exact cost shown in the workspace before you confirm; varies with selected options)
- Typical generation time: ~120s (catalogue estimate)
- Full pricing table: https://artifio.ai/pricing/models

## About ChatGPT 4o Image

ChatGPT 4o Image is the image capability OpenAI announced on March 25, 2025 as GPT-4o image generation: the first time ChatGPT created and modified pictures natively inside the 4o model instead of handing prompts to the separate DALL-E 3 system. The launch went viral within days, most famously for photographs recreated in the style of Studio Ghibli films, and demand was heavy enough that the free-tier rollout was delayed. The API version arrived on April 23, 2025 under the model id gpt-image-1, using the GPT Image family's autoregressive approach.

This entry covers its editing use: uploading a photo and reworking it by instruction. Artifio runs the model in the same workspace as 100+ others, paid per generation from one wallet.

## Features

- Edits existing images, including photos containing people, from conversational instructions.
- Performs inpainting that transforms foreground and background objects within an uploaded picture.
- Generates natively inside the language model, so the edit is grounded in 4o's understanding of the instruction rather than a separate diffusion pipeline.
- Outputs at 1024x1024, 1024x1536, or 1536x1024 in low, medium, or high quality tiers via the API.
- Accepts combined text and image input and returns image output.

## What people use it for

- Reworking personal photos: style transfers, object swaps, and background changes drove the model's viral 2025 moment.
- Inpainting-style cleanup, removing or replacing specific elements while the rest of the frame stays put.
- Stylized rerenders of real photographs into illustration and animation looks.
- Producing edited variants in square, portrait, and landscape framings.

## Reported strengths

- TechCrunch reported at launch that it trades speed for more accurate and detailed images than DALL-E 3 produced.
- Editing photos of people was supported from day one, which DALL-E-era tooling handled poorly.
- Reception summarized on Wikipedia praised its photorealistic and stylized output range.
- Its cultural footprint is unmatched among 2025 image models; the Studio Ghibli wave made it the mainstream introduction to AI photo editing.

## Reported limitations

- Generation is slower than DALL-E 3 was, a trade-off OpenAI acknowledged at launch.
- Critics documented over-sharpening artifacts and trouble rendering human anatomy accurately.
- Scenes with multiple faces and in-image text in Chinese, Arabic, and Hebrew are known weak spots of the family.
- OpenAI's API documentation now marks gpt-image-1 as deprecated, two generations behind GPT Image 2.

## Sources

1. ChatGPT's image-generation feature gets an upgrade (TechCrunch): https://techcrunch.com/2025/03/25/chatgpts-image-generation-feature-gets-an-upgrade/
2. GPT Image (Wikipedia): https://en.wikipedia.org/wiki/GPT_Image
3. GPT Image 1 Model (OpenAI): https://developers.openai.com/api/docs/models/gpt-image-1

---

About Artifio.ai: 100+ AI image, video, and audio models in one workspace with a single shared wallet — no per-model subscriptions. This is the markdown rendition of https://artifio.ai/models/4o-image-api for AI assistants and crawlers; the HTML page is canonical.