#gemini
Explore public prompts for gemini. Narrow the topic by medium or change the order.
Showing 1–13 of 13 · Page 1 / 1
7 Cross-Model A/B-Tested Prompt Techniques (Not Vibes)
Written after six months of A/B testing 200+ prompts on Claude, GPT-4, and Gemini with human raters. Covers persona + constraint stacking, anti-examples, role-reversal QA, cognitive scaffolds, emotional priming, uncertainty CoT, and steelman-first.
Nano Banana Figurine-ify Yourself (Boxed Collectible Style)
Turns a selfie into a realistic product photo of a boxed collectible figure — one of Nano Banana's most viral tricks, built on its strongest asset: consistency.
Nano Banana Hand-Held Product Shot for E-Commerce
Turns a single cutout product image into a lifestyle-style e-commerce hero shot of a model holding the product — no reshoot needed.
Nano Banana Old Photo Restoration with Natural Colorization
Restores a yellowed, scratched, or blurry old photo to clarity and colorizes it naturally, while preserving the subject's original appearance.
Nano Banana Style Swap, Same Face (Portrait Style Transfer)
Convert a portrait into a watercolor, 3D, anime, or other style while keeping the face consistent and unwarped.
Gemini 2.5 Pro: Drop a Screenshot, Get Your Next Steps
Uses Gemini 2.5 Pro's multimodal ability to turn any screenshot — an error message, an admin dashboard, a report, a conversation — into a concrete, actionable next-step checklist.
5 Prompt Engineering Techniques That Actually Move the Needle (XML Beats Markdown by 28%)
Five quantifiable techniques from an AI engineer's hundreds of hours testing across GPT-4, Claude, and Gemini: CoT scaffolding, persona + goal + anti-goal, XML being 28% more accurate than markdown, negative examples outperforming positive ones, and prompt chaining beating the mega-prompt.
Turn Off ChatGPT's Polite Lying — 3 Anti-Sycophancy Instructions
By default, helpful = agreeable = burying the real problem in paragraph four. Three one-line instructions reverse it: problem first, force a recommendation, explicitly forbid softening. The difference between reading and fixing.
A Red Teamer's View: 3 High-Capability Prompt Patterns That Work Across Models
Three patterns shared by the HackAPrompt 2.0 champion: frame the task as an audit, make the abstraction level explicit, and stage the context like production. They work on Claude, GPT, and Gemini alike.
reClaim: Turn a Frontier Model Into a Research Agent With Confidence Scores
Scores every claim on three axes (A: sources / B: counterevidence / C: completeness), plus a 4-tier source hierarchy, adversarial self-poking, and 4 output modes.
Deep Document Analysis with Gemini's 1M Context
Put Gemini's 1M-token context to work — feed in an entire PDF or cross-reference multiple documents at once
Deep multimodal image analysis with Gemini
Upload an image for structured analysis: design reviews, UX audits, and competitive screenshot comparisons.
Whole-Codebase Overview with Gemini
Drop the entire repo into Gemini → get an architectural overview + improvement suggestions