Fitness prompt tested across four image models
The author ran the same photorealistic fitness prompt through GPT Image 2.5 Sunburst, GPT Image 2.5 Flare, Nano Banana Pro and Seedream 5.0 Pro, and asks which handled it best.
Just a fitness prompt I tested with GPT Image 2.5 Sunburst, GPT Image 2.5 Flare, Nano Banana Pro and Seedream 5.0 Pro. Which one do you think handled it better? Prompt: 2x2 grid, AI infers extreme fitness subjects: INPUT: $ SUBJECT ACTOR = infer_lifter($ SUBJECT) // mid-movement, strained EFFORT = infer_exertion($ SUBJECT) // sweat, vein, flushed skin, clenched jaw GEAR = infer_equipment($ SUBJECT) // rack, dumbbell, cable, chalk MIRROR = infer_gym_mirror($ SUBJECT) // smudged, fingerprinted, sticker corner LIGHT = infer_fluorescent($ SUBJECT) // flat overhead, greenish, unflattering MOTION = infer_rep_motion($ SUBJECT) // bar path blur, shaking FLAWS = infer_flaws($ SUBJECT) // motion smear, noise in shadows Φ = sweat_specular(LIGHT, EFFORT) // highlight sheen on skin Compose: candid gym capture or mirror shot, harsh fluorescent light, sweat and effort visible, motion blur on the moving load. Photoreal, 9:16 vertical.