How to make an AI hug video

Steps

1. 1

   Open the AI Hug Generator

   Go to the AI Hug Generator tool page. The studio there is set up for hug videos, so there is no prompt to write. You can go through the settings before you sign in.

2. 2

   Add one or two photos

   Under Photos, add one photo with both people, or one photo of each person (two at most). If you are not signed in, a sign-in box opens when you upload: create an account with an email and password, or continue with Google. Your settings are kept.

3. 3

   Pick a hug style

   Choose a warm, long hug (the default), a joyful hug with a little spin, a gentle side hug or a tearful reunion hug.

4. 4

   Choose a model

   Kling 3.0 Standard is selected by default. You can switch to Kling 3.0 Pro, Seedance 2.0 Fast, Seedance 2.0, MiniMax Hailuo 2.3, Veo 3.1 Lite, Veo 3.1 Fast, Wan 2.7 or Grok Imagine Video in the model menu.

5. 5

   Set the aspect ratio and length

   Vertical 9:16 is preselected on every model that offers it. Kling makes 5 or 10 second clips, Veo 4, 6 or 8 seconds, and MiniMax Hailuo 2.3 only 16:9 clips of 6 or 10 seconds. Longer clips cost more.

6. 6

   Press Generate

   The button shows the credit price first. With two photos, they are combined into one frame with Nano Banana before the video model animates the hug, and the price on the button covers both steps.

7. 7

   Download or share the video

   The clip appears in the results panel when it is ready. Download the MP4, or share it to the public gallery if you want to. Nothing is shared unless you choose to.

Tips

• To hug your younger self, upload a childhood photo and a recent photo of yourself as the two photos.
• Photos where both people face the camera with similar framing combine into the most natural frame.
• Run a scratched or faded old photo through the Photo Restorer first, then use the restored copy here.
• One photo with both people skips the combine step and costs less.
• Start at the shortest length and go longer once you like the motion.

Troubleshooting

The message says "The model's safety filter rejected this request"

The model provider refused the request, and the credits are refunded up to a daily limit. Veo 3.1 Lite refuses some real-person photos, so switch to Kling 3.0 Standard and try again.

A Seedance 2.0 Fast video fails with a photo of a real person

Seedance 2.0 Fast rejects start frames with realistic faces. The failed job is refunded like any other; use Kling 3.0 Standard for photos of people.

The message starts with "This request was blocked"

Our own check stopped the request before it started, so no credits were taken. The link in the message explains the rule.

Black bars at the top and bottom of a Veo video

Veo 3.1 Lite letterboxes a start frame whose shape differs from the video. Crop the photo to the aspect ratio you picked before uploading, or use Kling 3.0 Standard.

The upload says "File is too large" or "Unsupported file type"

Use a PNG, JPEG or WebP under the size limit shown in the message. A scan of an old photo is often very large, so export a smaller JPEG first.

The generation failed

The credits are refunded automatically and the result card says so. Try again or pick another model.

Questions