Head to head: Gemini Omni Flash 1.1 Image to Video vs PixVerse Lipsync
Gemini Omni Flash 1.1 Image to Video vs PixVerse Lipsync
By Ryan Merket · Published · Updated
A head-to-head across rainy street scenes, a water-balloon burst, and a ramen-camera move tests how well each video model follows specific visual instructions. The comparison focuses on prompt details, motion, and continuity across sampled frames.
Gemini’s clips more often held onto the requested scene and action: it repeatedly showed the Rialto News kiosk, subway grate, coffee-carrying woman, leaves, and pigeons, and its ramen shots delivered a visible orbit where PixVerse often stayed nearly static. At the Bakery 17 scene, Gemini better conveyed the rainy dawn and courier easing toward a stop; PixVerse’s warm glow was appealing, but its cyclist often kept riding past. The water-balloon sequence was the closest contest. Gemini usually made the balloon burst legible, with a spreading sheet and droplets, while PixVerse sometimes produced a more convincing slow-motion splash. In one of the three sampled trials, PixVerse took that task; in another, its splash read more like water hitting a surface than a balloon bursting. The aggregate scores were 30.3 for Gemini and 22.3 for PixVerse, and the task-win tally was 11–1 with no task ties. Even so, the saved result for this matchup is a tie, with an editorial strength of **Too close to call (not statistical)**. Those numbers summarize this saved task sample; they are not a basis for claiming broader model superiority. **Final call: Tie on this task sample. Gemini was more consistent on scene details and camera movement; PixVerse had a standout fluid-dynamics result, but not enough to change the saved verdict.**
Subway grate pigeons
A short continuous 16:9 video clip at a downtown corner outside the fictitious Rialto News kiosk: a burst of dry gingko leaves and paper ticket stubs gets pulled into a spinning column by air surging from a subway grate just as nine pigeons hop into it, lifting, wobbling, and re-forming into a loose flock around a woman in a silver trench coat who laughs and shields her coffee; the camera performs a dramatic low orbiting move clockwise around her and the grate, staying in one shot as late-afternoon sun flashes between passing buses and the particles and birds behave convincingly in the turbulent airflow, creating a lively, slightly chaotic city mood.
OpenAI: GPT-6 Luna: Model A more clearly shows the Rialto News corner, a woman holding coffee, a subway grate, swirling leaves, and pigeons, with a more consistent scene across the sampled frames. Model B has striking warm light and energetic birds, but the grate changes into a trash bin and the kiosk signage is garbled, weakening continuity and prompt adherence. (Second judge pass, order swapped — scores are the average of both: Model A more clearly matches the Rialto News setting, silver trench coat, coffee, subw | Anthropic: Claude Fable 5.1: Model A delivers the Rialto News kiosk, silver…
Fluid & particle dynamics
Cinematic slow-motion of a water balloon bursting, the sheet of water expanding and droplets scattering through the air in convincing detail against a dark background, hard side light, 16:9.
OpenAI: GPT-6 Luna: Model A clearly shows a red water balloon bursting into a stretching water sheet and scattered droplets against a dark, dramatically lit background. Model B has attractive slow-motion water dynamics, but depicts a splash rising from a surface rather than a balloon bursting. (Second judge pass, order swapped — scores are the average of both: Model A clearly shows the balloon before it bursts and then captures the expanding water sheet and scattered droplets against a dark background. Model B has | Anthropic: Claude Fable 5.1: Model A shows a clear red water balloon pierced…
Camera motion control
A single smooth orbit (roughly 180 degrees) around a steaming bowl of ramen on a wooden table, the camera circling at a constant speed and height in one continuous move, warm restaurant light, 16:9.
OpenAI: GPT-6 Luna: Model A shows a clear, continuous orbit around the bowl, with the restaurant background and ramen perspective shifting substantially while the warm setting remains consistent. Model B is visually polished, but the camera is nearly static and the lighting is less consistently warm. (Second judge pass, order swapped — scores are the average of both: Model A shows a clearer, steady orbit with substantial background parallax while keeping the bowl at a consistent height and scale; Model B is attract | Anthropic: Claude Fable 5.1: Model A clearly executes a wide orbit around th…
Dawn bakery calm
A short continuous 16:9 video clip on a narrow city side street at 5:58 a.m.: a bicycle courier in a mustard rain shell coasts slowly past the steamed-up window of Bakery 17, then eases to a stop as the baker inside flips on a row of warm pendant lights one by one, the amber glow spreading across the wet pavement while a blue dawn drizzle fades; the camera makes a gentle sideways dolly with a slight inward arc around the courier, matching the unhurried pace, and the mood should become quietly serene through the motion, pacing, and evolving light rather than any posed action.
OpenAI: GPT-6 Luna: Model A closely matches the named bakery, 5:58 a.m. blue drizzle, mustard-clad courier, and wet pavement, with the courier easing into a stop as the warm interior becomes more prominent. Model B is polished and smoothly composed, but lacks the Bakery 17 detail and the courier continues past rather than stopping. (Second judge pass, order swapped — scores are the average of both: Model A more clearly establishes the specified 5:58 a.m. Bakery 17 setting and shows the courier easing to a stop, wit | Anthropic: Claude Fable 5.1: Model A matches the prompt closely: a narrow si…
Matchup powered by OpenRouter.