Head to head: LTX 2.5 Image to Video Fast vs MiniMax (Hailuo AI) Video 01
LTX 2.5 Image to Video Fast vs MiniMax (Hailuo AI) Video 01
By Ryan Merket · Published
This matchup tests whether polished temporal consistency can outweigh precise execution of complex motion, occlusion, and lighting prompts. Across repeated task families, one model proved substantially more dependable.
LTX 2.5 Image to Video Fast wins decisively: **30.9 to 23.7**, with seven task victories to MiniMax’s one and a **limited confidence** statistical verdict. This was not a narrow aesthetic preference; LTX repeatedly executed the defining action of each prompt while MiniMax often produced attractive footage that missed the requested event. The clearest separation came in choreography and change over time. LTX handled the tram-depot scenes with more believable traffic, braking, wet reflections, and subject movement; it also made the teal coach pass fully behind the fogged glass tunnel and re-emerge. MiniMax preserved details such as Route 41 and sometimes offered cleaner subject separation, but its coach never completed the crucial occlusion, and its tram motion and worker choreography diverged from the brief. Lighting was equally one-sided. LTX visibly moved from warm sunset into cool dusk while activating the lamp, whereas MiniMax largely held on an already-lit, nearly static scene. LTX also won one cathedral-glide test through stronger spatial progression and atmospheric dust. MiniMax’s sole victory came in the other cathedral run, where its architecture, centered movement, and temporal consistency were cleaner; LTX’s candles and pews fluctuated more, and its lighting-transition shots showed some framing and furniture drift. **Final call: LTX 2.5 Image to Video Fast is the clear winner. MiniMax can look polished, but LTX is far more reliable when the prompt demands specific motion, evolving light, and coherent occlusion—not merely a stable image in motion.**
Tram Depot Crossing Swirl
At blue-hour drizzle outside the Kestrel Quay tram depot, a short continuous 16:9 shot begins with a slow lateral dolly to the right along a striped service crossing as seven different subjects move independently at once: a lime maintenance scooter zips left to right, two depot workers in orange rain capes wheel separate tool carts in opposite directions, a cream-and-maroon tram eases forward from the shed, three cyclists thread through the foreground at different speeds, and a courier van noses in from frame right then brakes; wet rails mirror the sodium yard lights, tiny umbrella reflections jitter in puddles, and the mood is tense but everyday, with each person and vehicle keeping distinct believable motion and spacing without blending, warping, or borrowing limbs or wheels from one another.
OpenAI: GPT-5.6 Sol Pro: Model A more faithfully depicts the striped depot crossing, forward-emerging tram, three cyclists, independently moving workers and scooter, and braking van with believable spacing and atmospheric realism. Model B is temporally coherent and visually striking, but its tram travels laterally, the workers move similarly rather than oppositely, and the highly saturated look reduces realism. (Second judge pass, order swapped — scores are the average of both: Model A better captures the striped crossi | Anthropic: Claude Fable 5.1: Model A delivers nearly every prompt eleme…
Single continuous shot
One unbroken take gliding slowly through a candlelit cathedral from the entrance toward the altar, no cuts, jumps, or transitions, dust and warm light in the air, 16:9.
OpenAI: GPT-5.6 Sol Pro: Model A clearly progresses from the open entrance toward the altar in a visually coherent glide, with strong dust-filled light rays and warm candlelight. Model B is attractive and consistent, but the sampled views show less obvious forward movement and less airborne dust or atmospheric light. (Second judge pass, order swapped — scores are the average of both: Model A more clearly depicts a continuous forward glide from the entrance toward the altar, with convincing spatial progression, airborne | Anthropic: Claude Fable 5.1: Model A clearly executes the brief: a conti…
Lighting transition
A living room at dusk as the sun sets over a few seconds: the warm light visibly dims and shifts to cool blue while a lamp flickers on, the transition smooth and believable, locked-off frame, 16:9.
OpenAI: GPT-5.6 Sol Pro: Model A clearly depicts a locked-off, smooth progression from warm sunset light to cool dusk as the lamp activates and brightens, with strong visual continuity. Model B is polished and stable but remains nearly unchanged across the sampled frames, with the lamp already on and no convincing lighting transition. (Second judge pass, order swapped — scores are the average of both: Model A clearly depicts a believable progression from warm sunset to cool blue dusk, with the lamp switching on and incr | Anthropic: Claude Fable 5.1: Model A clearly executes the full transiti…
Coach Behind Glass Tunnel
In a single continuous 16:9 shot at first light, the camera makes a gentle forward push down the platform of the remote Bracken Hollow coach stop as a bright teal intercity coach labeled Route 41 glides from left to right, passes fully behind a long fogged glass pedestrian tunnel that hides it for a moment, and then re-emerges on the far side unchanged in shape, color, markings, and trajectory; a woman in a mustard parka on the near platform keeps pace briefly with a rolling silver suitcase, gulls lift from the roofline, cold pink sunrise edges the metal beams, and the mood is quiet and expectant, with the coach remaining perfectly coherent through the occlusion.
OpenAI: GPT-5.6 Sol Pro: Model A closely depicts the teal coach crossing behind the fogged tunnel and re-emerging coherently, while also including the mustard-coated woman, suitcase, gulls, and pink dawn atmosphere. Model B is visually polished and consistent but has the coach traveling away beside the glass structure rather than moving left to right through the required full occlusion. (Second judge pass, order swapped — scores are the average of both: Model A more closely depicts the teal coach traveling left to right | Anthropic: Claude Fable 5.1: Model A delivers the core choreography: a…
Matchup powered by OpenRouter.