How 20 AI image models render: “A macro still photograph of an extraordinary, mythical creature constructed entirely from layered, realistic autumn leaves. The creature is posed gracefully on a moss-covered fallen log deep within a serene, misty forest floor during early morning. Its entire form is seamlessly sculpted from diverse foliage: its body and back are layered with overlapping, crisp scarlet maple leaves, its slender legs are formed from slender, serrated golden birch leaves, and its elegant, curved tail is crafted from a single, elongated amber fern frond. The creature features expressive, glossy black obsidian stone eyes that capture sharp, detailed reflections of the surrounding trees. Every individual leaf shows immaculate, crisp physical textures—from the microscopic, branching network of translucent veins catching the soft morning sun to the weathered, curling edges and tiny dew droplets resting on the surfaces. The background is a soft-focus.”

Same prompt, sent verbatim, at each model's default settings and a fixed seed. Judge: could this be a photograph? Skin, light, lens behaviour. 196 community votes so far.

Community prompt by @anyafoxx · voting closed 2026-11-12

20 text-to-image models rendered this photorealism prompt with the identical text at their default settings and a fixed seed. GPT Image 2.6 Flare is the community favourite so far, winning 72% of its 18 head-to-head votes on this prompt. Qwen Image 3.0 was the fastest at 6.7s.

What to look for: believable skin and fabric, physically plausible light and shadow, lens behaviour such as depth of field and motion blur, and no waxy or over-smoothed surfaces.

FAQ

Which AI model renders this prompt best?

GPT Image 2.6 Flare leads this prompt in the BudgetPixel AI Image Arena, winning 72% of its 18 blind head-to-head votes. Vote on the arena page to add your judgement.

Were the prompts adjusted per model?

No. Every model received the exact same prompt, unmodified, at its default settings, with no negative prompt or style suffix. Some models rewrite prompts internally, and that is part of what is being compared.

Can I run this prompt myself?

Yes. Each tile’s “Try” link opens the BudgetPixel workshop with this prompt and that model preselected.