Omni Flash is dangerously close to becoming the model you outgrow before you master it. And I say that as someone who just burned a ridiculous amount of credits pushing it way past what I thought it could do. I wanted to see if I could use Omni Flash to build a realistic UGC-style beard grooming video, not just a guy talking to camera, but an entire grooming routine with continuity. Same character. Same bathroom. Same facial structure. Different tools. Different grooming stages. Trimming. Carving. Razor detailing. Product application. Final reveal. And getting this result wasn't one prompt. I generated. It failed. I regenerated. It almost worked. I changed the keyframe. It broke something else. I burned more credits. I tried again. That's the part people don't see when they look at the final 78 seconds. AI video isn't just about having access to a powerful model. It's knowing how to direct the model when it refuses to give you what you want. Omni Flash can produce some genuinely impressive realism. But that's exactly why I think Google needs to move faster. The competition isn't waiting. We're getting to a point where “good enough” isn't good enough anymore. If Omni Flash wants to stay relevant in this model battle, the ceiling needs to keep moving. For now, though… I wanted to find out where that ceiling was. So I pushed. And this is what came out. Made with Omni Flash 1.1, powered @FlowbyGoogle .
A man grooms his beard in a bathroom, trimming and then conditioning it, in a UGC-style routine.
- 8.3K views
- 125 likes
- 16 reposts
- 33 replies
- 58 bookmarks
This video is Chuzzy‼️| AI Creative Director’s work, shown from their public post. See it on X.
