ChatGPT Images 2.6: Create GIFs From A Simple Sketch

• GIF creation. ChatGPT can turn a prompt, photo, or sketch into a short looping GIF.
• Saved reference images. You can save a character, product, or person once and reuse it in later chats without uploading it again.
• Better consistency. OpenAI says character, product, and brand details stay more consistent across edits.

Choose the prompt for your task: create an image, edit an existing one, or animate it. Replace the bracketed placeholders with your own details.

Starting Image Prompt

Create a clean, reusable visual of [subject / mascot / product] for [audience and use]. Keep these details fixed: [shape, colors, outfit, proportions, distinctive features]. Use [style and brand palette]. Keep the background simple and avoid small text.

Clearly define what you want to create, the audience, and details ChatGPT should not alter. See our CODO framework for more tips on structuring your request.

Targeted Edit Prompt

Keep the subject, proportions, colors, outfit, camera angle, and composition unchanged. Change only [one specific area or detail]. Do not redesign anything else. Compare the result with the original before making another edit.

Naming one change keeps the request focused. Check the result for unintended changes elsewhere.

GIF Prompt

Animate only [one simple movement]. Keep the face, text, logo, product shape, and camera fixed. Make a short seamless loop. If the visual contains a real person, keep the portrait static and move only surrounding elements.

One simple movement makes it easier to check the loop for flicker or distorted features.

1. Turn an illustration into a GIF

Start with a sketch, illustration, or brand mascot. A mascot could wave in a welcome message, a butterfly could flap its wings, or a product illustration could show one moving part.

The example below shows a butterfly sketch and the colored version ChatGPT generated.

To animate it, use the GIF prompt above and ask for gentle wing movement while keeping the body and background still.

Top: the sketch you start with. Bottom: the colored version ChatGPT generated from it.

2. Create variations of one approved visual

Reuse a visual for different messages while keeping the speaker, brand colors, and other key details consistent.

For the Lead with AI PRO graphic example below, the variations include a session promotion, a question-led post, and a summary of the main ideas.

Use the targeted edit prompt to change a headline or another specific detail. If you want a new layout, remove the instruction to keep the composition unchanged. Check that the portrait, names, and logos still match the original.

One PRO visual adapted into three variations.

Other uses include product images with seasonal backgrounds, a mascot in different campaign graphics, and event graphics resized for email, presentation slides, or social posts.

• Check the starting image to ensure details are correct. Confirm names, logos, colors, and text before creating more versions.
• Review every edit and watch the full loop. Look for changed lettering, distorted features, or a visible jump when the animation restarts. Use our DRAFT Check to review the result.
• When you create animations with text, logos, or portraits, ask ChatGPT to generate MP4 videos instead for sharper outputs. Keep GIF formats for simple, short loops.

Plans and Availability

Images 2.6 is available on all plans, in ChatGPT and Codex, on the web, desktop, and mobile.

GIF creation is rolling out gradually, so it may not appear in your account yet. Check back if you don't see it.