Sam Altman showcases what images users can create with ChatGPT Images 1.5: Sketches, 3D dolls, plush toys and more

OpenAI CEO Sam Altman shared young inside information well-nigh ChatGPT Images on wed, an upgraded range feature powered by OpenAI’s latest image generation model. The San Francisco-based company says the tool makes it easier to create new images or edit existing photos, with results that better match what users imagine.

Sam Altman showcases new image tools in ChatGPT

In a post on Wednesday, Altman showed how the new Images tab in ChatGPT can be used to create fun visuals. The examples included turning photos into 3D glam dolls, ornaments, sketches, plush toys, doodles and holiday portraits.

The image he shared also showed more ideas users can try. These include making a holiday card, fixing and restoring old photos, recreating yourself in the style of famous paintings, turning images into keychains, designing album covers and even asking ChatGPT to suggest a matching outfit.

Altman described the feature as a simple and enjoyable way to create images directly inside ChatGPT.

Images 1.5 now available for all users

Altman also announced that Images 1.5 has launched in ChatGPT and is available through OpenAI’s API. He said the update brings better image quality, faster results and new editing tools. To mark the launch, he shared a playful image of himself dressed as a firefighter on a December calendar, created using the new feature.

OpenAI claims ‘better photo editing’

According to OpenAI’s blog, the new model is better at understanding what users want. It can make more useful photo edits, create more realistic clothing and hairstyle try-ons, and apply different styles while keeping the main details of the original image.

The model can handle many types of edits, such as adding or removing objects, mixing images together and changing their layout, without losing what makes the image special.

Easy creative changes

OpenAI says the tool is also strong at creative changes, like adding text, changing layouts or reimagining an image in a new style. These features work for both simple and more detailed ideas.

Many of these options are available through ready-made styles in the ChatGPT Images section, so users do not need to write long prompts. OpenAI says the model follows instructions more accurately than earlier versions, making it easier to get the result you want.

Can Images 1.5 be an answer to Gemini Nano Banana Pro?

After Google added major image editing upgrades to Gemini in August under the playful name Nano Banana, OpenAI appears to be answering back with Images 1.5. The new update brings stronger image editing tools directly into ChatGPT, making visual changes more flexible and controlled.

A key improvement is the ability to edit only selected parts of an image while leaving the rest unchanged. Users can add or remove objects, or adjust colours and styles, without the image losing its original look or turning into something completely different. This kind of precise editing was one of the highlights of Gemini’s earlier update.

ChatGPT now also supports combining multiple images into a single scene, another feature already seen in Gemini. Together, these upgrades suggest OpenAI is aiming to match and challenge Google’s progress in AI-powered image editing, while keeping the tools simple to use inside ChatGPT.

---

---