Upload a photo, character, or illustration and turn it into an animated GIF with AI. Bring your original image to life or use it as a visual reference to create a new scene, pose, or action.
Explore these animated GIF examples, then create your own.
Upload an image, describe how it should move, then select Generate GIF.
Failed generations return the credits used.
An image to GIF maker turns a single still photo or illustration into an animated GIF. Upload one image and describe the movement you want: AI uses the image and your instructions to generate a moving result you can preview and download. You can bring the original composition to life, or use a recognizable person or character as a visual reference for a new scene.
Explore text and image creation options and more animation examples with our AI GIF Generator.
Both approaches start with one image. Your prompt tells AI whether to keep the original scene or build a different one around your reference character.
Upload a photo or illustration and describe a small change: a blink, a wave, drifting clouds or a moving reflection. Ask AI to preserve the main subject, background and framing. The goal is a moving version of your original image; fine details can still change during generation.
Treat the uploaded person or character as a visual reference. Describe a different environment, pose or action, and name the features to keep: face, hairstyle, outfit, colors and illustration style. AI generates a new composition guided by the reference. Character consistency is a goal, not an exact identity lock.
Start with one JPG, PNG or WebP image up to 8 MB. Choose a clear subject with enough space around the area you want to move. A visible face helps with a portrait; a full-body reference gives AI more information for gestures or changes in pose.
Decide whether to animate the original scene or use the subject as a character reference for a new setting or action. Describe that choice directly in your prompt. Try a starting prompt from the two approaches above or the example cards, then edit it to fit your image.
For a photo, say what moves and what stays still. For a new scene, describe the environment and one main action, then list recognizable features to preserve. Be explicit about keeping or replacing the background. A fixed camera helps keep attention on your subject.
Choose a style, Flare or Sunburst, and an aspect ratio. Use Auto style when you want the reference to guide the look. Resolution, frame count and timing are set automatically.
Review the 24-credit cost, confirm that you have permission to use the image, and click Generate GIF. If prompted, sign in and click Generate GIF again. Watch the completed animation in History, check the faces and motion, then download it. Each new generation uses credits.
Prefer to skip the reference image and start directly with an idea? Create an animation with Text to GIF.
Compare the original image with the actual animated GIF. Each case includes its creative prompt, what changed and the measured output details.
What changed: A proud step becomes a wide-eyed stumble, a squash and a springy recovery.
What was preserved: Yellow chick, teal overalls, orange feet, dark outline style and white background.
Use the uploaded chick illustration as the sole character and style reference. Preserve its cream-yellow feathers, three head tufts, orange beak and feet, teal overalls, round body and dark outlines in every frame. CORE JOKE: trying to look cool after an embarrassingly tiny stumble. Start in the relaxed reference pose, pause, then let the chick puff its chest and take one extremely confident step. It trips over its own foot, eyes popping wide and wings windmilling briefly as the whole round body squashes down. Catch its balance with one springy upward rebound, freeze in an exaggerated innocent I-meant-to-do-that grin, then settle back to the original pose. Exactly two wings and two feet, no extra limbs or props. One readable comic action with anticipation, strong expression contrast, gentle squash and recovery. Square 1:1, pure white background, fixed camera, full character centered with generous margins. About 2–3 seconds, naturally repeating loop. No text, captions, logos or watermark.
What changed: A composed smile turns into suppressed giggles, a forward lean and a hand-over-mouth laugh before recovery.
What was preserved: Black hair, round glasses, lavender sweater, photographic appearance and gray backdrop.
Use the uploaded portrait as the sole identity and photographic-style reference. Preserve the adult woman’s shoulder-length black hair, round glasses, lavender knit sweater and recognizable face in every frame. CORE JOKE: trying very hard not to laugh in a serious conversation. Begin composed with a polite smile. Press the lips together as the cheeks puff slightly, eyebrows rise and eyes begin to crinkle. Her shoulders then shake in two restrained little giggles; she quickly brings one hand to cover her mouth, leans forward with an unmistakable delighted laugh, catches herself and sits upright trying to look serious again. End in the original composed pose. Keep realistic facial anatomy, exactly two hands, rigid glasses and natural hair movement; no melting or extra fingers. The whole upper body follows the laughter. Clear buildup, burst of emotion, comic recovery, naturally joined loop. Square 1:1, unchanged light gray background, fixed camera, centered upper body with safe margins. Aim for 3–4 seconds with genuinely evolving movement, not a long still hold. No text, captions, props, logos or watermark.
What changed: A calm pose turns into a head tilt, eager bounce and flapping ears, followed by a disappointed slump.
What was preserved: Tricolor fur, floppy ears, yellow bandana and ivory studio setting.
Use the uploaded beagle photo as the sole animal and photographic reference. Keep the same brown, black and white fur markings, floppy ears, yellow bandana, body proportions and ivory background in every frame. CORE JOKE: the word treat instantly overrides a dog’s attempt to look dignified. Begin sitting with a very serious calm face. Suddenly the ears perk and head tilts hard to one side, eyes brighten; the tail accelerates into an eager wag. The dog rises into one little excited front-paw bounce, ears flapping naturally, leans forward hopefully, then notices there is no treat and slowly slumps back into the original dignified sitting pose with a funny disappointed look. Return to the opening expression to join the loop. Four paws only, natural canine anatomy, realistic fur, stable scarf; no human hands, no food, no added props or captions. Square 1:1, fixed camera, full dog visible with room for the bounce. Aim for 3–4 seconds, continuously evolving action, brief readable emotional beats and a natural infinite loop.
What changed: The coffee ripples, swirls upward and releases a droplet as steam curls above the cup.
What was preserved: Coral mug, right-side handle, ivory tabletop, product framing and photographic style.
Use the uploaded coral-orange coffee mug photograph as the sole product and style reference. Keep the exact mug silhouette, handle on the right, color, tabletop, lighting and fixed three-quarter view recognizable throughout. THEME: this coffee has a little too much energy. Begin with the calm reference cup. Small concentric ripples spread across the dark coffee, then the liquid makes a tiny energetic swirling mound while a graceful ribbon of steam curls upward into a loose spiral. One small coffee droplet pops above the rim and falls neatly back into the cup, making a soft final ripple. The steam unwinds and the liquid settles back to the opening still surface. A soft highlight travels across the glaze during the action. Keep the ceramic rigid and stationary: no cartoon face, no eyes, no limbs, no cup deformation, no spilled liquid or added objects. One satisfying buildup, playful peak and smooth recovery. Square 1:1, ivory background unchanged, fixed camera, whole cup visible, ample margin above the steam. Aim for 3–4 seconds with multiple connected stages of movement and a natural infinite loop. No text, caption, logo or watermark.
What changed: A band of ripples crosses the lake, the boat rocks and the reflected sunrise shimmers before settling.
What was preserved: Mountain silhouettes, pines, shoreline, wooden boat, sunrise palette and wide composition.
Use the uploaded alpine lake photograph as the scene reference. Preserve the mountain silhouettes, shoreline, pine trees, sunrise colors, small wooden boat and overall composition throughout. Bring the calm landscape to life with a connected gentle sequence: a breeze sends a band of ripples toward the boat, the boat rocks softly and turns just a few degrees, reflected sunrise light breaks into sparkling streaks on the moving water, and a few pine branches sway while pink clouds drift almost imperceptibly. Let the wave pass and the boat settle back to its starting position as the water reflection calms. Keep mountains and shore fixed, natural photographic motion, no warping or rapid time-lapse. Square 1:1, fixed wide camera, no zoom, no people, no new objects, no text or watermark. Aim for 3–4 seconds of evolving water, boat and reflection movement with a natural infinite loop.
What changed: White background becomes a flower garden. The chick sniffs a daisy, startles at a butterfly and recovers with a sheepish grin.
What was preserved: Yellow chick design, teal overalls, orange feet and dark-outlined flat illustration style.
Use the uploaded chick as the sole character and illustration-style reference. Preserve its cream-yellow round body, three head feathers, orange beak and feet, teal overalls and thick dark-brown outline in every frame. Create a NEW SCENE: a simple pastel flower garden with a small daisy beside the chick, green grass, two rounded bushes and pale blue sky, matching the flat 2D illustration style. CORE JOKE: the tiniest butterfly causes the biggest startle. Begin with the chick proudly sniffing the daisy. A tiny pink butterfly flutters off the flower toward its beak. The chick goes cross-eyed, recoils with a huge startled open-beak expression, squashes then hops backward with both wings thrown up. Realizing the butterfly is harmless, it gives a sheepish grin, leans back toward the flower and returns to the starting pose while the butterfly returns to the daisy. One clear reaction, brief anticipation, expressive peak, natural body follow-through and gentle squash/rebound. Exactly two wings and two feet. Fixed camera, full character and flower visible with generous safe margins, square 1:1. About 3–4 seconds, evolving motion and naturally repeating loop. No text, captions, logos or watermark.
What changed: The mascot moves to a desk, types with exaggerated effort, flops forward and rebounds beneath the English caption “OUT OF ENERGY.”
What was preserved: Yellow chick, head tufts, teal overalls, dark outlines and flat illustration style.
Use the uploaded chick as the sole character and 2D illustration-style reference. Preserve cream-yellow feathers, three head tufts, orange beak and feet, teal overalls, round proportions and dark-brown outlines in all frames. NEW SCENE: a simple pastel home-office desk with one tiny laptop, one mug and pale mint wall, clear uncluttered flat illustration. CORE JOKE: acting busy for two seconds then completely running out of energy. The chick begins sitting upright behind the desk, wearing an excessively determined expression. It leans forward and types furiously with both little wings, eyes wide, body bouncing with effort. Suddenly its eyelids droop, it gives a huge silent yawn, then the round body slowly compresses and the head flops face-first onto the closed-clear area of the desk beside the laptop. Pause for a comic beat. It springs upright with an innocent alert smile and returns to the original determined pose. Keep exactly two wings, consistent laptop and mug, minimal props. Gentle squash, big expression contrast and naturally connected return. CAPTION: exactly "OUT OF ENERGY" at the top in bold rounded uppercase English lettering, navy fill, thick white outline. The words, letterforms, position and size must remain stable in every frame, no other text, character and props never cover caption. Square 1:1, fixed camera, whole upper body and desk visible, safe margins. Aim for 3–4 seconds with expressive action throughout, natural infinite loop. Deliver a real animated GIF, no watermark.
What changed: The standing character dances, turns around and blows a kiss before returning to her opening pose.
What was preserved: Pink twin buns, crescent hairpin, teal eyes, blue cardigan and chibi anime style.
The girl dances happily, spins around, blows a kiss, then naturally returns to her starting pose.
Output dimensions, timing, and file sizes shown here are measured from the downloaded GIFs.
Choose a source that already shows the details you care about. A clear reference gives the AI image to GIF process more useful information than a crowded or heavily obscured picture.
Use a well-lit face with visible eyes and a clear hairstyle. Avoid heavy occlusion and strong blur. Include shoulders and hands for a wave, or the full body for a standing gesture. A portrait cannot reliably supply details that are completely outside the original frame.
Choose one clearly visible animal, with recognizable fur markings and enough room around its ears or tail. Start with a slow blink, head tilt or tail movement. If the action depends on a paw or tail, use a reference that actually shows that part of the animal.
Use a clean silhouette, readable colors and consistent linework. A full-body character reference is useful for new poses and gestures. Name the design features to retain in your prompt, and keep Auto style if you want the original artwork to guide the result rather than requesting a different look.
Pick a sharp product image with a simple background and clear outline. A moving reflection or a little steam often needs less reinvention than a complete rotation. Inspect labels, logos and geometry after generation: AI may alter small lettering or packaging, even when you ask to keep them.
Look for a stable composition with one obvious moving element, such as water, clouds or leaves. Specify a fixed camera and name the landforms or buildings that should remain still. Subtle movement is a practical starting point when you want the photo’s original atmosphere and framing to remain recognizable.
For a photo to GIF prompt, describe the movement first, then specify what should stay recognizable. Name the facial features, hairstyle, clothing, object shape or colors that matter. Keep the original background and a fixed camera when you want the existing scene to come alive. Give the action a clear beginning, reaction and return to its starting pose.
For a character reference GIF, describe the new setting and one connected action sequence. Then list the face, hairstyle, outfit, proportions and visual style to preserve. Say explicitly that the background or pose should change. Match the new environment to the emotion or joke, and leave enough room around the character for its movement.
Describe the feeling or joke, then let one main action carry it: anticipation, a clear emotional peak, a brief hold and a recovery into the opening pose. Ask the body to follow naturally, with a little squash and rebound where it suits the style. Use only essential props. Choose a plain background for a reaction sticker, or name a new setting when it supports the joke.
Specify the exact short phrase in quotation marks. Ask for bold rounded lettering at the top, navy fill, white outline and a stable position across frames. Leave space above the subject. AI can misspell or distort lettering, so check the actual animation; this is a prompt instruction, not a separate text editor.
Use a sharp, readable subject. Crop away distractions while leaving enough room for the movement.
Give the action a beginning, a surprising reaction and a recovery. A pause followed by an exaggerated expression can be more readable than several unrelated gestures.
Name the important face, hair, outfit and color details. Review the finished frames; instructions cannot guarantee exact consistency.
Say “keep the original background and a fixed camera” for subtle photo motion. Describe the new environment explicitly for a reference-based scene.
Try a small action before a complex pose change. Ask the subject to return to the starting pose, and check the loop join in the actual GIF.
An image to GIF maker uses one still photo or illustration and a written prompt to create an animated GIF. Here, AI generates motion from your reference. You can keep the original scene or describe a new one using the same recognizable person or character. It is not a traditional tool for combining a sequence of photos.
Upload one image and describe the motion or new scene. Choose a style, Flare or Sunburst, and an aspect ratio. Review the 24-credit cost, confirm permission and click Generate GIF. If prompted, sign in and click Generate GIF again. Preview the animation in History before downloading.
Yes. One photo is the reference input; you do not need to upload a sequence of frames. AI generates new animation frames based on the photo and your instructions. A clear subject and a small, focused action are useful starting points. Uploading several images to assemble a slideshow is not supported by this tool.
Upload one JPG, PNG or WebP image up to 8 MB. Use a clear image that you own or have permission to use. The reference is processed on the server and sent to the generation provider when you generate. Avoid uploading sensitive images that you are not comfortable sharing for that processing.
Yes. Ask for a natural blink, a slight smile, a head turn or a gentle wave. Name the facial features, hair and clothing you want to preserve, and say whether the background should remain unchanged. AI can still introduce differences, so inspect the face, hands and motion in the finished GIF.
Yes. Use the uploaded image as a character reference and describe a different setting, pose or action. Tell AI which features should stay recognizable, such as the hairstyle, outfit, colors and illustration style. This is a prompt-based approach to generating a new composition; it does not lock a face or guarantee exact character identity.
Yes. Reuse the same reference image and change the requested action or environment for each generation. Keep your description of the character’s defining features consistent. Each generation uses credits and can vary in appearance. Review the separate results before treating them as a consistent character series.
Yes. Upload a clean illustration or cartoon reference, then describe one action. Auto style lets the source image and prompt guide the appearance. Name the linework, palette, silhouette and proportions to preserve. More elaborate gestures or a different visual style can change the character, so review the generated frames.
The reference and your prompt guide recognizable features, but preservation is not exact or guaranteed. Faces, hair, clothing, fingers and small details may change. A clear source and modest motion can make the task more focused. Check the actual result, and avoid relying on generation for identity-critical or exact product representations.
You can request both in a character-reference prompt. Describe the new environment and one clear action, then specify the facial features, hairstyle, clothing and style to keep. Do not also ask to preserve the entire original composition if your goal is a different scene. The output may still vary from the reference.
You can guide the result in words: for example, “move the clouds, keep the mountains still, fixed camera.” This editor does not offer motion masks or precise per-region controls. AI interprets the instructions, so unintended movement can appear. Start with one moving element and inspect the result.
Animating the original image aims to keep the scene and composition while adding movement. A character-reference GIF uses the subject’s recognizable appearance to guide a different setting, pose or action. Both start with one uploaded image. Describe your intended approach in the prompt, or use a starting prompt from the two approaches or example cards.
Download an animated GIF from the result panel after generation. Check the downloaded file for its actual pixel dimensions.
Yes. Download a completed GIF from the result panel. Check its animation, dimensions and file size before sharing it, since platforms have different upload limits. You can also download the editorial examples below their case cards. Generated results are not guaranteed to have a perfect loop or exact reference preservation.
New accounts receive 22 free signup credits. Each Image to GIF generation costs 24 credits, so you need additional credits before your first image generation. You can use the signup credits to try Text to GIF instead. The editor shows your balance and generation cost.
Start with one image. Bring its scene to life or give your reference character a new setting and action.
