One GPT Image 2.5 prompt turns any photo's lower half into an office-supply collage illustration
Fresh-Resolution182 · reddit · 2026-09-20
A Reddit user shared a full GPT Image 2.5 prompt that keeps the original photo in the top half and rebuilds the bottom half as a hand-drawn, mixed-media collage made of office supplies — sticky notes, paperclips, tape, receipts, envelopes and stamps.
Key prompt design choices:
- The model first distills the photo's meaning, emotion and symbolism, then rebuilds it rather than copying object by object
- Top half gets only a light high-end color grade; the subject is never stretched or altered
- Large deliberate whitespace is treated as an active compositional element, with subjects off-center, hugging edges or partly cropped
- Watercolor, colored pencil, paper grain and handwritten notes keep an editorial, handmade feel
- The most emotionally charged colors from the photo are reorganized into a light palette
An example shows a sailboat and iceberg at dusk. The prompt is also packaged as a skill that batch-processes whole folders of photos, available at GitHub repos nevertoday/xxd-panel-225 and AtlasCloudAI/awesome-gpt-image-2.5-prompts.
More from Multimodal
- Free Local HTML Prompt Randomizer With Weights, Consistency Locks and Batch Variations — klartreumer · 2026-09-20
- GPT renders Opus's poetic imagining of Claude: a timeless oracle with fourteen eyes — repligate · 2026-09-20
- YuE2 tips: KSampler ~16x faster, CFG sweet spots, LoRAs and 900s song fix — Diligent-Rub-2113 · 2026-09-20
- XGEN-JING trends on Hugging Face: egocentric world model with joint audio-video generation — XGENlabs · 2026-09-20
- AI Short Film 'It Happens to Every Wizard' Goes Viral on Reddit — Sweet_Composer_1323 · 2026-09-20
- Original AI Sci-Fi Short Film 'Slug' Asks What If Brainpower Could Be Sold — Luna_Nguyenn · 2026-09-20