ChatGPT adds native GIF generation powered by Images 2.6

Native animation in the chat

OpenAI is simplifying the process of making custom animated GIFs by building the tool directly into ChatGPT. Previously, creating these files usually required users to open separate editing software or manually stitch together individual image frames. Now, the process is streamlined into a single interface.

The new feature is powered by an unannounced model called Images 2.6. This model allows the AI to take simple text descriptions or uploaded photos and transform them into smooth, looping animations. It represents a shift toward making motion generation more accessible to casual users.

This launch follows the company's decision to shut down its standalone Sora AI video generation app in March. While GIFs are less complex than full-scale video projects, putting them inside the main chat window makes creating motion content much more effortless for the average person.

Precise creative controls

The new GIF generator provides several tools for creators to refine their animations. Users can easily tweak the playback speed of a generated loop to get the timing just right. The interface also allows for frame-by-frame viewing, which helps in checking the consistency of the motion.

These controls are designed to let users make quick edits without ever leaving the conversation window. By keeping these tools native, OpenAI aims to reduce the need for third-party apps. This is particularly useful for creating quick meme loops or simple visual assets.

The company recently launched a sticker creation tool within ChatGPT as well. These new features combined suggest a broader push toward making ChatGPT a more flexible destination for everyday visual content creation.

Models for different needs

OpenAI is making the Images 2.6 model available to people using ChatGPT and Codex on web, desktop, and mobile apps. To support different types of workflows, the company is offering specialized versions for developers. These models build on the dual-release approach first introduced with the Images 2.5 update last September.

Developers can now choose between two specific models depending on their project goals. The GPT-Image-2.6 Flare model is designed to handle tasks where speed is the primary concern. In contrast, the GPT-Image-2.6 Sunburst model is built for projects that require a higher level of visual quality.

By providing dedicated API tools, OpenAI is ensuring that both high-volume automated tasks and high-fidelity creative projects can utilize the new technology. This tiered approach helps different types of users get the specific performance they need from the underlying AI.

Rapid growth in image volume

The expansion of these features comes as OpenAI sees a significant increase in how much content people are making. The company states that the total number of images produced by users across their platforms and developer API now exceeds 4 billion every week. This is a major jump from the 3 billion weekly images reported back in September.

This growth highlights the increasing demand for AI-generated visuals in both personal and professional settings. As the tools become more capable and easier to use, the volume of output continues to climb to new heights every month. The native GIF tool is expected to contribute to this upward trend.

The move toward more integrated tools like GIF generation and stickers reflects a strategy of making the platform more versatile. By offering more ways to create and manipulate media, OpenAI is positioning ChatGPT as a comprehensive hub for visual communication.

What to expect next

OpenAI is gradually rolling out these updates to its various platforms. Users on web, desktop, and mobile applications should see the Images 2.6 features appearing in their accounts over time. This rollout ensures that the new capabilities reach the widest possible audience steadily.

The focus on looping animations and stickers suggests that OpenAI may continue to refine short-form motion content. While full video generation remains a separate space, the integration of GIFs into the chat window makes casual animation a standard feature. This makes it easier for creators to experiment with motion without a steep learning curve.

As the Images 2.6 model becomes more established, we can expect more refinements to the editing controls. The goal appears to be providing a seamless experience where a user can go from a text prompt to a finished, high-quality animation in just a few seconds.