This is an AI video generation comparison for
image-to-video 
prompt:
Man comically gobbling up spaghetti directly with his mouth. Surreal, comical....
Log in to see full prompt.
Tested: April 21, 2025
Tested: May 10, 2025
Image to video with ltxv-2b-0.9.6 distilled in ComfyUI
Tested: May 10, 2025
Text prompt placed at the end of the storyboard.
Tested: May 10, 2025
Kling 1.6 image to video = simple prompt
Tested: May 10, 2025
Pika Turbo img-2-video
Tested: May 10, 2025
V4 model at 360p. Prompt part about 'sucking it in' kept triggering moderator so was removed.
Tested: May 10, 2025
S2.0 Pro (Seaweed alpha image to video)
Tested: May 10, 2025
I2V-01 optimisation enabled
Tested: May 10, 2025
Looking good, pretty funny facial expression.
Is the man actively engaging with the spaghetti using only his mouth (no utensils)?
Does the man appear to be mid-bite or mid-suck, not just posing near the plate?
Is his facial expression exaggerated or humorous to match a comical tone?
Is the spaghetti visibly abundant and chaotic, enhancing the comedic excess?
Do strands of spaghetti appear to be moving into his mouth (not static)?
Check out the results from FramePack (FramePack) vs LTX Video by Lightricks (LTX Video 2b-0.9.6 Distilled) vs Sora vs Kling AI (Kling 1.6 Standard) vs Pika (Pika Turbo) vs PixVerse (PixVerse V4) vs Dreamina AI (Seaweed Alpha) vs Hailuo AI x Minimax vs MAGI (Magi-1) vs GROK (Grok Imagine v0.9) for similar or identical prompts side-by-side.
Surreal flamingo teacup overflow
Wild blooms by alpine lake camera pan right
