Illustration & anime
How 22 AI image models render: “Cel-shaded anime illustration of an adult rooftop beekeeper tending lively hives among marigolds and herbs at dusk, with bees drifting past distant windows. Palette of saffron, violet, mint, and midnight blue, crisp contours and gentle gradients.”
Same prompt, sent verbatim, at each model's default settings and a fixed seed. Judge: line quality, colour, style consistency. 128 community votes so far.
Fresh prompt from the daily stream · voting closed 2026-10-07
22 text-to-image models rendered this illustration & anime prompt with the identical text at their default settings and a fixed seed. Grok Imagine Image 2 is the community favourite so far, winning 80% of its 10 head-to-head votes on this prompt. Grok Imagine Image Quality was the fastest at 5.1s.
What to look for: confident line work, consistent style from edge to edge, deliberate colour and readable shapes.
- 80%22.5s · 50 cr Try
- Recraft v4 Pro67%Recraft · retired from voting16.3s · 300 cr Try
- GPT Image 264%OpenAI63.4s · 42 cr Try
- FLUX 3 Image63%Black Forest Labs20.9s · 32 cr Try
- GPT Image 2.5 Sunburst62%OpenAI33.0s · 40 cr Try
- ImagineArt 1.5 Pro56%ImagineArt · retired from voting51.9s · 55 cr Try
- Muse Image 1.056%Meta13.8s · 12 cr Try
- Seedream 5.0 Flash55%ByteDance29.1s · 22 cr Try
- GPT Image 2.5 Flare50%OpenAI24.7s · 40 cr Try
- ImagineArt 250%ImagineArt32.8s · 60 cr Try
- Qwen Image 3.0 Pro47%Alibaba10.8s · 36 cr Try
- Krea 2 Large43%Krea31.2s · 80 cr Try
- Seedream 5.0 Pro43%ByteDance73.6s · 49 cr Try
- Ideogram v4.540%Ideogram14.8s · 80 cr Try
- Nano Banana 2 Lite38%Google9.8s · 36 cr Try
- Ideogram v431%Ideogram15.6s · 40 cr Try
- Qwen Image 3.027%Alibaba6.7s · 32 cr Try
- 5.1s · 80 cr Try
- Kling v3 Omni15%Kuaishou74.7s · 35 cr Try
- Krea 2 Medium15%Krea14.8s · 40 cr Try
- Recraft v4Recraft · retired from voting13.7s · 45 cr Try
FAQ
- Which AI model renders this prompt best?
- Grok Imagine Image 2 leads this prompt in the BudgetPixel AI Image Arena, winning 80% of its 10 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.