"Hand grabs bag" video Prompt Generation Comparison

This is an AI video generation comparison

A purple handbag with a gold chain rests on a white pedestal, set against a lime green backdrop. A hand with glossy red nails elegantly enters the frame from the right, fingers brushing the gold clasp before gently lifting the bag. As the chain softly clinks, the camera tracks smoothly in a close-up, following the bag as it's carried out of frame with flair.

Generated: April 20, 2025


Appears to be unable to remove an object from scene in this case. 2 generations attempted: hand grabs but fails to lift

Generated: April 22, 2025


Image to video, done in Vidu's Q1 model. Same issue as in Hailuo Minimax: 1-1=1, something that supposed to diminish or disappear form old place when taken into a new place stays there. The grabbing movement is great though.

Generated: April 25, 2025


Kling 1.6 perfect prompt execution. Not enough time to complete the whole move at slower speed it's taken but we can see bag being lifted, shadows underneath it.

Generated: April 25, 2025


Well that's an unexpected hand! So 'hand with glossy red nails' became also a red hand, why not? At least this AI knows 1-1=0, bag doesn't also remain on shelf once picked up.

Generated: April 25, 2025


I2V-01 model by Hailuo still can't count. Almost finished the right way, then decided to spawn the bag into its original spot somehow.

Generated: April 25, 2025


Wow, Pika 2.2 model truly excelled on this one. Proper hand and fingers, grabbing and removing, with shadows underneath AND the bag is gone!

This prompt tests:

Product Presentation & Material Rendering: High-gloss purple leather, gleaming gold chain, luxe texture, rich surface reflections, pristine pedestal contrast.

Color & Composition Contrast: Vibrant lime green background, white pedestal, bold red nails—intentional palette clash for visual impact.

Hand Interaction & Object Handling: Manicured hand enters smoothly, fingers brush gold clasp, natural grip and lift, elegant motion.

Check out the results from FramePack, Vidu AI, Kling AI, MAGI-1, Hailuo AI x Minimax, Pika for similar or identical prompts side-by-side. Note generation dates and comments about models used.