Choose the role before the photo
Start by deciding which half of the joke belongs to you. On the table casts your photo as the person who mistakes the small round surface for a seat. Walking in casts you as the guest who understands the furniture. Both of us adds a second photo field so a friend can join the scene. Keep one subject in each image: a group shot leaves the intended role unclear.
Give the studio a clean starting point
Use one subject per JPG, PNG or WebP image, up to 10 MB per file. A clear, evenly lit face or a full-body pet photo gives the model a useful reference. Select On the table or Walking in for one photo, or Both of us for two separate photos. Only use images you own or have permission to share.
Generate, follow the task and download
Sign in, choose 5, 10, or 15 seconds, select 768p, 1K (1080p), or 2K and review the framing. The scene prompt is already filled in. Submitting uploads your photos and sends the generation request to Metaso MiniMax-H3. Follow the saved task in My videos and download the finished result. If submission is uncertain, resume that task instead of starting another.
Choose 5, 10, or 15 seconds and 768p, 1K (1080p), or 2K. If submission is uncertain, continue the saved request instead of starting another.