Latest News

ChatGPT now lets you generate GIFs

A crude drawing of an owl next to a more detailed illustration of a cartoon owl with its wing raised in a fist.

ChatGPT will now allow users to make GIFs from text prompts or uploaded images. The new feature is powered by Images 2.6, OpenAI‘s latest AI image generation model, unveiled on Wednesday.

Rolling out globally to ChatGPT and coding tool Codex today, Images 2.6 enables users to either describe the GIF they would like it to generate or upload an image to be animated. Users can also upload a rough drawing, and have the AI model to generate a more complete version to animate.

Once generated, The Verge reports that ChatGPT’s new feature will allow users to adjust the GIF’s animation speed, as well as view and edit individual frames. OpenAI stated in a since-deleted X post that users can also have Images 2.6 save their uploaded pictures to be reused as prompts in the future.

A screenshot of a deleted X post by @ChatGPT announcing Images 2.6.

Credit: Mashable screenshot: @ChatGPT via X

Significantly, Images 2.6’s GIF-generating capability is not an AI video generator. Instead, OpenAI suggested that Images 2.6 can be used to create animated storyboards or to “explore character movement before video production.” OpenAI discontinued its AI video generation app Sora earlier this year. The Wall Street Journal reported that it had few users and lost the company approximately $1 million per day.

OpenAI appears to have largely focused its visual AI generation models on still images since then. However, the launch of Images 2.6 suggests it may be flirting with moving pictures once more. Last month, OpenAI launched Intelligent UI, a new ChatGPT feature that allows users to generate interactive visual interfaces.

Two images of an orange suitcase in water, side by side. One is generated by GPT‑Image‑2.5 Flare and the other by GPT‑Image‑2.6 Flare. Text below indicates that the former was generated in 7.8 seconds, while the latter was generated in 4.82 seconds.

Credit: OpenAI

According to OpenAI, Images 2.6 also generates images 50 percent faster than its previous model, Images 2.5. Generation is still likely to take a moment, though. The company says that its speed-focused GPT‑Image‑2.6 Flare model can generate an image in about six seconds, provided its quality is set to low. OpenAI’s GPT‑Image‑2.6 Flare and GPT‑Image‑2.6 Sunburst models are rolling out in the company’s API.

Images 2.6 is available now across all of ChatGPT and Codex’s subscription tiers.


Disclosure: Ziff Davis, Mashable’s parent company, in April 2025 filed a lawsuit against OpenAI, alleging it infringed Ziff Davis copyrights in training and operating its AI systems.

Show More

Related Articles

Leave a Reply

Your email address will not be published. Required fields are marked *

Back to top button

Adblock Detected

Please consider supporting us by disabling your ad blocker