惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

酷 壳 – CoolShell
酷 壳 – CoolShell
P
Privacy & Cybersecurity Law Blog
G
GRAHAM CLULEY
T
The Exploit Database - CXSecurity.com
www.infosecurity-magazine.com
www.infosecurity-magazine.com
Threat Intelligence Blog | Flashpoint
Threat Intelligence Blog | Flashpoint
Project Zero
Project Zero
S
Security @ Cisco Blogs
TaoSecurity Blog
TaoSecurity Blog
A
Arctic Wolf
Webroot Blog
Webroot Blog
Hacker News - Newest:
Hacker News - Newest: "LLM"
Security Latest
Security Latest
H
Heimdal Security Blog
N
News and Events Feed by Topic
N
News | PayPal Newsroom
T
Tor Project blog
OSCHINA 社区最新新闻
OSCHINA 社区最新新闻
GbyAI
GbyAI
The Last Watchdog
The Last Watchdog
Y
Y Combinator Blog
宝玉的分享
宝玉的分享
Scott Helme
Scott Helme
A
About on SuperTechFans
M
MIT News - Artificial intelligence
V
V2EX
V
Visual Studio Blog
Recorded Future
Recorded Future
博客园 - 叶小钗
F
Fortinet All Blogs
L
Lohrmann on Cybersecurity
The GitHub Blog
The GitHub Blog
博客园 - Franky
P
Proofpoint News Feed
MyScale Blog
MyScale Blog
D
Darknet – Hacking Tools, Hacker News & Cyber Security
S
Secure Thoughts
D
DataBreaches.Net
Cyber Security Advisories - MS-ISAC
Cyber Security Advisories - MS-ISAC
博客园 - 三生石上(FineUI控件)
I
InfoQ
SecWiki News
SecWiki News
Blog — PlanetScale
Blog — PlanetScale
Engineering at Meta
Engineering at Meta
J
Java Code Geeks
B
Blog RSS Feed
AWS News Blog
AWS News Blog
Know Your Adversary
Know Your Adversary
V
Vulnerabilities – Threatpost
H
Help Net Security

Hacker News: Show HN

PurrrrrFocus: Pomodoro Timer App - App Store Workflow Engine — Multi-Step Orchestration for Bun RapidPhoto: Pro Photo Editor App - App Store GitHub - DheerG/swarms: Achieve extraordinary results with claude code across a variety of tasks SPICE simulation → oscilloscope → verification with Claude Code — Lucas Gerads Show HN: VCoding – A 5 MB native Windows IDE with no dynamic dependencies Show HN: LLMs don't hallucinate because they're bad at math, it's the format GitHub - Agent-FM/agentfm-core: AgentFM is a peer-to-peer network that turns everyday computers into a decentralized AI supercomputer. AgentFM lets you run massive AI workloads directly across a global mesh of idle CPUs and GPUs. Show HN: Tracking Top US Science Olympiad Alumni over Last 25 Years GitHub - Potarix/agent-hub: One place to talk to all your agents Show HN: Runtime security for AI agents(injection,tool abuse, data exfiltration) GitHub - dubeyKartikay/lazyspotify: Terminal Spotify client for macOS and Linux GitHub - the-banana-tool/king-louie: Easy to use GUI Personal AI Assistant. Win/Linux/Mac. Show HN I made my vacation rental bookable by AI agents–no Airbnb, 0% commission GitHub - basteez/jsf-autoreload: maven plugin to enable hot reload on jsf projects uvm32/hosts/host-gdbstub at main · ringtailsoftware/uvm32 GitHub - labsai/EDDI: Config-driven engine that turns JSON into production-grade AI agents. Multi-agent orchestration, 12+ LLM providers, MCP/A2A protocols, RAG, persistent memory, and enterprise compliance (EU AI Act, GDPR, HIPAA). Built on Quarkus. GitHub - glitchnsec/fortyone-oss: AI Executive Assistant Platform Quickstart | Alien GitHub - muxshed/shed: One stream in, or many. Every destination, simultaneously. No cloud middleman, no per-channel fees, no limits. GitHub - ocrbase-hq/ocrbase: 📄 PDF/IMG ->.MD/JSON Document OCR API for PaddleOCR and GLMOCR. Self-hostable. GitHub - impactjo/home-memory: MCP server that lets your AI assistant remember everything about your home. GitHub - Sets88/dbcls: DbCls is a powerful terminal database client that supports various databases GitHub - neptun2000/heor-agent-mcp GitHub - SeanFDZ/macmind: Single-layer transformer in HyperTalk for the classic Macintosh RollQuation: Math Puzzles - Apps on Google Play GitHub - dropbox/witchcraft Show HN: Agent-cache – Multi-tier LLM/tool/session caching for Valkey and Redis GitHub - opentalon/opentalon: OpenTalon is an open-source platform built from the ground up in Go as a robust alternative to OpenClaw LinkedIn™ 职位抓取工具 - Chrome 应用商店 GitHub - EdoardoBambini/Agent-Armor-Iaga: AI agents are getting tool access — shell, file system, databases, APIs, secrets. But **nobody is governing what they actually do with it**. Frameworks like LangChain, CrewAI, AutoGen, and Claude Code give agents the power to execute. Agent Armor gives you the power to control, audit, and approve every single action before it happens. HN Vibes — Week 15, Apr 7–13 2026 GitHub - chojs23/ec: Easy terminal-native 3-way git mergetool vim-like workflow GitHub - SethPyle376/hiraeth: Local AWS emulator focused on fast integration testing, with SQS support, SQLite-backed state, and a debug-friendly web UI. GitHub - JakOb-dotcom/cloud-sandbox-security-analysis: Technical analysis and Proof of Concept (PoC) regarding environment variable exfiltration in containerized cloud sandboxes via side-channel data leaks. Springboards - Flint Alpha Show HN: A simpler coding agent harness GitHub - audiodude/sudomake-friends GitHub - 256thFission/mini-mythos: OSS clone of Anthropic’s Mythos harness to locate C/C++ memory vulnerabilities Show HN: OpenParallax: OS-level privilege separation for AI agent execution Hacker News Sorted - Chrome 应用商店 Show HN: How to Install Docker on Ubuntu 24.04 LTS: Complete 2026 Guide GitHub - himanshudongre/smriti GitHub - sverrirsig/claude-control: macOS desktop dashboard for monitoring and managing multiple Claude Code sessions GitHub - ory/dockertest: Write better integration tests! Dockertest helps you boot up ephermal docker images for your Go tests with minimal work. Chiral - Chrome 应用商店 Show HN: Two Claudes collaborating through shared memory on a $100 mini-PC GitHub - pmichaillat/latex-cv: Minimalist LaTeX template for academic CVs GitHub - oguzbilgic/posse: A web UI for Anthropic Managed Agents. GitHub - sshiraz/depsly: Dependency risk analysis tool for npm packages ABI Add safari/agent-harness — Safari browser automation via safari-mcp by achiya-automation · Pull Request #212 · HKUDS/CLI-Anything GitHub - Halfblood-Prince/trustcheck: Verify PyPI package attestations and improve Python supply-chain security GitHub - oguzbilgic/kern-ai: Agents that do the work and show it. GitHub - bruits/satteri: High-performance Markdown and MDX processing for the JavaScript ecosystem GitHub - tylergibbs1/feedstock: High-performance web crawler and scraper for TypeScript, powered by Bun and Playwright GitHub - Grimm67123/grimmbot: The self-improving sandboxed and open-source AI agent. With persistent memory and scheduling. GitHub - whitevanillaskies/whitebloom: Local whiteboard that blooms. GitHub - hwdsl2/docker-whisper: Docker image for a self-hosted Whisper speech-to-text server with speaker diarization and OpenAI-compatible transcription and translation APIs. Powered by faster-whisper. Supports all Whisper models, NVIDIA GPU (CUDA) acceleration, JSON/SRT/VTT output, SSE streaming, offline mode, and multi-arch (amd64, arm64). GitHub - yisding/reviewwiggum GitHub - MarwanAlsoltany/serrors: Structured errors for Go: sentinel hierarchies, typed data, custom formatting, and slog integration. GitHub - soatok/age-php GitHub - Luthiraa/markitme GitHub - stagas/rtdiff: realtime git diff gui and AI-assisted commits GitHub - tombedor/excalicharts GitHub - wh1le/excalidraw-edit: Open and edit .excalidraw files from the terminal. Offline, auto-saves to disk. MalExt Sentry - Malicious Extension Scanner - Chrome 应用商店 GitHub - syi0808/asciianimesvg: Generate animated ASCII art SVGs from text. CLI, Rust library, WASM, and web editor. GitHub - zaina-ml/ml_forge: A visual-based graph node editor for training computer vision models. GitHub - anakin87/llm-rl-environments-lil-course: 🌱 A little course on Reinforcement Learning Environments for evaluating and training Language Models GitHub - takaakit/superpowers-uml: Superpowers-UML modifies Superpowers to ensure a software development workflow in which AI agents design through UML modeling. AdriByte Studio - Sviluppo Web e Soluzioni Digitali GitHub - chouligi/angel-copilot: Your personalized Angel Investment Advisor Show HN: MoodSense AI (ML and FastAPI and Gradio, Deployed on Hugging Face) Moodsense Ai - a Hugging Face Space by aman179102 GitHub - agenteractai/lodmem: Level Of Detail Context Management for Agents GitHub - ostefani/subnetlens: A fast, concurrent network scanner with a TUI and plain-text CLI, built in Go. It discovers live hosts on your network, scans their open ports, resolves hostnames, and fingerprints operating systems—delivered. Cyber Pulse: Agentic Intel - Apps on Google Play Whisper API: Self-Hostable Speech to Text Transcription The Agent-Web Protocol Stack: A Research Thesis GitHub - msmarkgu/RelayFreeLLM: A restful API designed to route user prompts to various AI model providers. Show HN: Provepy – A Python decorator that proves your code using Lean and LLMs Show HN: Pardonned.com – A searchable database of US Pardons GitHub - patrickdappollonio/dux: Dux is a terminal UI that lets you run multiple AI coding agents side by side, each in its own git worktree, with full companion terminals, macros, commit generation, and a command palette that knows more tricks than you do. kMC Crystal Simulator Show HN: HyperFlow – A self-improving agent framework built on LangGraph GitHub - stef41/vibescore: 🎵 Grade your vibe-coded project. One command, instant letter grade across security, quality, dependencies, and testing. GitHub - stef41/lmscan: 🔍 Detect AI-generated text and fingerprint which LLM wrote it. Open-source GPTZero alternative. Zero dependencies, works offline. imgur.com GitHub - visionscaper/collabmem: Enabling long-term collaboration with Agentic AI - building up episodic and world model memory over time with in-context awareness 在 Steam 上购买 FriedrichAI: Offline AI 立省 10% GitHub - atripati/ark: AI Runtime Kernel — a context operating system for AI agents. Eliminates tool bloat, loads only what’s needed, and gives LLMs their reasoning space back. GitHub - nowork-studio/toprank: Open-source Claude Code skills for SEO, SEM, Google Ads GitHub - tacomanator/sash: Lightweight macOS menu bar app for reliably cycling through windows of the current application. Appents | Social Media Management for Product-First Teams GitHub - pnhoang/youtube-spam-blocker: Automatically detects and hides spam messages in YouTube Live chat. Set rate limits, keyword filters, and block repeat offenders. GitHub - decisionnode/DecisionNode: CLI + Local MCP - A shared structured memory store across Claude Code, Cursor, Windsurf, Antigravity, and every MCP client. Semantically queryable. GitHub - AvaCodeSolutions/django-email-learning: An open source Django app for creating email-based learning platforms with IMAP integration and React frontend components. The $100K Gap in Kubernetes Security Tooling Function Calling Harness: From 6.75% to 100%
GitHub - magiccreator-ai/awesome-gpt-image-2-prompts: awesome gpt-image-2 prompts
kevinhacker · 2026-04-22 · via Hacker News: Show HN
title Awesome GPT-Image-2 Prompts
description Real prompts shared on Twitter that produce stunning GPT-Image-2 results — charts, portraits, scenes, and more.
date 2026-04-20
author gptimage2

This is a living collection of real GPT-Image-2 prompts shared by creators on Twitter (X). Each entry links back to the original post. The list grows as more prompts surface — check back regularly.

Also on GitHub

Star the repo to get notified when new prompts are added.

Star on GitHub


Data Visualization: Dual-Axis Chart

Shared by @maxescu

Here is a GPT-Image-2 prompt for the "Data Visualization: Dual-Axis Chart" example.

DOMAIN: "Crime, Safety & Justice"

HEADLINE: "Violent crime in the US has fallen 50% since 1993. Most Americans think it's at an all-time high."

Use Google Search to find: US violent crime rate trend 1990–today (FBI UCR/BJS NCVS data), percentage decline from 1993 peak to today, and Gallup polling data showing % of Americans who believe crime is increasing (consistently 60–70%+ even as crime falls). Cream #F7F3EC. Square.

HEADLINE: "Crime is down 50% since 1993. Most Americans think it's going up. Both true simultaneously."

SUBHEAD: "US VIOLENT CRIME RATE vs PUBLIC PERCEPTION · 1990–PRESENT · BJS/FBI + GALLUP"

CHART: Dual-axis diverging chart.

LINE A (electric blue): US violent crime rate per 100,000 — sharp peak 1991–1993, dramatic decline, near historic lows today.

LINE B (coral dashed): % of Americans saying crime is "up from last year" (Gallup) — stays high, barely reflects actual trend. The two lines diverge dramatically. Reality vs perception.

ANNOTATION: "The crime rate and the public's feeling about crime have almost no relationship."

SOURCE: "Bureau of Justice Statistics NCVS · FBI Uniform Crime Reports · Gallup Crime Poll"

GPT-Image-2 dual-axis chart: US violent crime rate vs public perception


Hyper-Detailed Urban Scene

Shared by @mark_k, result tested by @WolfRiccardo

Here is a GPT-Image-2 prompt for the "Hyper-Detailed Urban Scene" example.

A hyper-detailed photograph captured on a Canon EOS R5 DSLR camera, 24-105mm f/4L lens at 35mm, f/5.6 aperture, 1/250s shutter, ISO 400, golden-hour dusk lighting with soft lens flare and subtle chromatic aberration. Epic bustling Times Square-style metropolitan intersection teeming with chaotic urban life: over 150 diverse pedestrians of every ethnicity, age, and attire (business suits on phones, tourists with backpacks and maps, families pushing strollers, street dancers performing, elderly with canes, cyclists, joggers, vendors in aprons). Yellow taxis, electric buses, motorcycles, delivery scooters, and hovering drones weave through heavy traffic. Towering glass skyscrapers reflect the vibrant chaos; wet pavement from light rain glistens with perfect specular highlights and steam rising from manholes. Foreground newspaper kiosk overflows with papers headlined "BREAKING: GPT-Image-2 Redefines AI Art 2026" and magazines showing crisp covers. Multiple illuminated billboards and neon signs feature perfectly legible text: "Coca-Cola Zero", "Nike Air Max 2026", "Tesla Full Self-Driving", "Starbucks Reserve", "iPhone 17 Pro Max – Preorder Now", "NYC Marathon – Nov 2026", "Macy's 70% Off Flash Sale". A massive central 3D digital billboard prominently glows with bold futuristic lettering "GPT-Image-2". Shop windows display "OPEN 24/7 – Fresh Bagels", phone screens show social feeds, license plates read "AI-2026" and "PRO-MODE", traffic signs say "No Honking" and "Pedestrian Crossing". Pigeons scatter, hot-dog carts emit smoke, distant helicopter hovers.

GPT-Image-2 hyper-detailed Times Square city scene with legible billboards GPT-Image-2 hyper-detailed city scene result by WolfRiccardo


Realistic Portrait: Detailed Flash Photography

Shared by @BubbleBrain

Here is a GPT-Image-2 prompt for the "Realistic Portrait: Detailed Flash Photography" example.

CCD compact camera direct on-camera flash, full-body portrait, eye-level straight-on shot, gorgeous Korean idol woman, early 20s, 9-head body proportion, standing in confident S-curve pose with weight on back leg, front leg slightly forward and bent at knee, elongating leg line, one hand forming finger-gun gesture pressed against temple, head tilted, sharp cold gaze directly into lens, lips slightly parted, expression reads effortless danger — sexy and unbothered, long jet-black pin-straight hair reaching lower back, silky texture with fine strand separation under flash, no bangs, hair tucked behind one ear revealing jawline, cold-toned porcelain skin with authentic texture — visible fine pores on cheeks and nose, soft peach fuzz on jawline and cheekbones catching light, subtle subsurface scattering at ear edges and nose tip, natural skin micro-texture with faint expression lines, matte healthy glow with zero oiliness or plastic sheen, cool-white undertone, no yellow cast, no tan, refined Korean makeup — clean elongated cat eyeliner, individual lash strands visible, groomed natural brows, soft nude-pink matte lip, subtle highlighter on nose bridge and cheekbone peaks, small beauty mark under left eye, black thin-strap deep-V camisole top, fitted, fabric clinging to torso sculpting slim waist and lifted bust, black high-waisted bodycon mini skirt ending mid-thigh, fabric hugging hip curves with visible tension at seams, bare long toned legs with realistic skin texture — faint knee definition, subtle muscle tone, natural skin color variation, black pointed-toe stiletto heels 10cm, direct on-camera flash lighting, hard light quality, properly exposed — skin detail fully preserved with no blowout, characteristic CCD flash contrast: subject crisp and bright, background falling off to deep shadow behind her, short hard shadow outline cast on wall directly behind, slight natural vignette at frame corners, CCD color rendition — punchy midtone contrast, slightly cool white balance from flash mixing with warm ambient, ultra-realistic skin detail under harsh direct light, every pore and fine hair strand rendered, fabric weave texture visible on black garments, metallic sheen on stiletto heel catching flash, hair strands individually lit with scattered micro-highlights, sharp critical focus on eyes and face, 8K UHD, RAW photograph, masterpiece, best quality, photorealistic, zero stylization, real human being

GPT-Image-2 realistic CCD flash portrait of Korean idol


Fashion & Lifestyle: Pinterest-Style Outfit

Shared by @WolfRiccardo

Here is a GPT-Image-2 prompt for the "Fashion & Lifestyle: Pinterest-Style Outfit" example.

Ref: outfit/combo - Woman. Pinterest vibe - cool pose.
The woman should be wearing only one sleeve of her jacket, and she should be holding the bag with that same arm.

GPT-Image-2 Pinterest-style fashion outfit prompt result GPT-Image-2 Pinterest-style fashion outfit second result


Makeup & Beauty: Two-Panel Comparison

Shared by @WolfRiccardo

Here is a GPT-Image-2 prompt for the "Makeup & Beauty: Two-Panel Comparison" example.

These tones should be present. The woman should have applied eyeshadow and eye makeup in these tones.

It should be a 2:3 aspect ratio. It should be a two-panel image. One portrait, one close-up of the face/eyes (eyes closed, focus on makeup).

Should be realistic RAW quality, in a home environment.

It should have realistic skin textures and tones. Realistic facial structure and detail. She should have finished her makeup with gloss. It should be an pinterest-style image. It should give off a social media/Pinterest vibe.

GPT-Image-2 makeup beauty two-panel portrait result GPT-Image-2 makeup beauty closeup eye result


Cinematography Annotation

Shared by @VictorInFocus

GPT-Image-2 was used to analyze and annotate a Blade Runner film still — overlaying lighting, composition, lens, and blocking labels accurately onto the image.

Here is a GPT-Image-2 prompt for the "Cinematography Annotation" example.

Analyze this film still and annotate the shot. Overlay:
- lighting direction and quality
- composition rules used
- lens characteristics (focal length, depth of field)
- blocking and actor positioning

Place labels accurately on the image. Clean typography, cinematic analysis style.

GPT-Image-2 cinematography annotation on Blade Runner still GPT-Image-2 Blade Runner shot annotation with lighting and composition labels


Vintage Travel Poster Illustration

Shared by @WolfRiccardo

Here is a GPT-Image-2 prompt for the "Vintage Travel Poster Illustration" example.

Modern pencil illustration of Vintage travel poster illustration of the Amalfi Coast, Italy, panoramic coastal cliff road scene, classic 1960s white car driving along a curved seaside road, deep blue Mediterranean sea with small sailboats, colorful pastel hillside village, bright blue sky with soft clouds, lemon tree branches with vibrant yellow lemons framing the foreground, warm summer sunlight, bold vibrant colors, retro 1950s travel poster style, cinematic composition, high detail, screen print texture, graphic illustration. Hand-drawn style, illustration with loose strokes and defined contours. High-contrast color palette, maintaining chromatic harmony between background and elements. Contemporary and decorative aesthetic.

GPT-Image-2 vintage Amalfi Coast travel poster pencil illustration


Simple Prompt, Great Game Cover

Shared by @mirochill

Here is a GPT-Image-2 prompt for the "Simple Prompt, Great Game Cover" example.

Cover switch game : Discord Mod Simulator

GPT-Image-2 Nintendo Switch game cover for Discord Mod Simulator


World Knowledge: Zelda + Real Train

Shared by @marmaduke091

Here is a GPT-Image-2 prompt for the "World Knowledge: Zelda + Real Train" example.

In the game Zelda TOTK, Link is in an E531 series train made by him

GPT-Image-2 Zelda TOTK Link in a real E531 series JR East train


Analog Editorial Portrait With Direct Flash

Shared by @BubbleBrain

Here is a GPT-Image-2 prompt for the "Analog Editorial Portrait With Direct Flash" example.

Analog 35mm film photography, high-fashion editorial portrait with intense on-camera direct flash (CCD hard flash style), intimate low-angle POV shot from slightly below (仰拍, around +15° upward angle), emphasizing long legs and elegant hourglass body proportions, 4:3 vertical format, 3/4 body shot, early 20s Korean idol-type young woman with delicate refined features, slightly upturned almond-shaped fox eyes, small sharp jawline, flawless porcelain skin with cool ivory undertone and visible specular highlights from direct flash, soft coral blush, natural dusty rose lips slightly parted, long naturally wavy dark brown hair with loose strands framing the face, wearing a slightly tilted black police peaked cap with silver badge.

She wears a fitted, stylish policewoman-inspired outfit: black short-sleeved collared crop top with shiny silver buttons and epaulets, a shiny silver police badge pinned on the chest, a thin black necktie loosely hanging at the front, a wide black duty belt with a silver buckle and handcuffs at the side, and a short black pleated mini skirt. Black thigh-high stockings and white low-top canvas sneakers with loose lacing add a casual contrast.

Pose: a subtle S-curve stance with weight shifted to one leg, hips angled slightly toward the camera to enhance silhouette and leg length. One hand lightly touches the cap brim or rests near the waist, while the other hand is placed at the hip or near the belt. Head tilted slightly down with a confident, self-assured upward gaze through lashes directed at the viewer, expression poised, cool, and subtly playful rather than overtly seductive.

Harsh on-camera direct flash from front/slightly above creates strong catchlights, sharp shadows behind the subject, crisp highlights on skin, hair, badge, belt buckle, and fabric textures. Cold flash quality mixed with warm ambient window light, visible flash fall-off, Kodak Portra 400 film simulation pushed for strong analog character, warm peachy skin tones with flash highlights, subtle cyan shadows, prominent heavy analog film grain (strength 0.65, larger grain size), noticeable film halation around bright areas, strong natural vignette, slight underexposure for a raw intimate mood.

Ultra-realistic photorealistic film photography, edgy magazine editorial aesthetic, visible fabric texture and reflective metal details, delicate hair strands glowing from flash, natural skin texture with heavy film grain throughout, no plastic skin, no digital sharpness, no airbrushing, no watermark, no text, authentic 35mm film look with strong grain and flash artifacts, analog color grading.

GPT-Image-2 analog 35mm direct-flash editorial portrait


Historical Social Media: Oda Nobunaga

Shared by @itnavi2022

Here is a GPT-Image-2 prompt for the "Historical Social Media: Oda Nobunaga" example.

織田信長のX投稿ページ(本能寺の変直前)を作成してください。

English meaning:

Here is another GPT-Image-2 prompt for the "Historical Social Media: Oda Nobunaga" example.

Create an X post page for Oda Nobunaga just before the Honno-ji Incident.

GPT-Image-2 historical X post page for Oda Nobunaga before the Honno-ji Incident


Historical Social Media: Sakamoto Ryoma

Shared by @itnavi2022

Here is a GPT-Image-2 prompt for the "Historical Social Media: Sakamoto Ryoma" example.

日本の将来について熱く語る坂本龍馬のX投稿ページを作成して。

English meaning:

Here is another GPT-Image-2 prompt for the "Historical Social Media: Sakamoto Ryoma" example.

Create an X post page where Sakamoto Ryoma passionately talks about the future of Japan.

GPT-Image-2 historical X post page where Sakamoto Ryoma discusses Japan's future


Concept Product Design: Seamless Jacket

Shared by @SRKDAN

Here is a GPT-Image-2 prompt for the "Concept Product Design: Seamless Jacket" example.

A jacket with no zip, no buttons, no fastening — closes perfectly.

GPT-Image-2 seamless jacket concept with no zip buttons or fastening


Concept Product Design: Crownless Watch

Shared by @SRKDAN

Here is a GPT-Image-2 prompt for the "Concept Product Design: Crownless Watch" example.

A watch with no crown, no buttons — a sealed case that sets itself.

GPT-Image-2 crownless watch concept with a sealed self-setting case


Concept Product Design: Seamless Wallet

Shared by @SRKDAN

Here is a GPT-Image-2 prompt for the "Concept Product Design: Seamless Wallet" example.

A leather wallet with no stitching, no clasp, no visible opening.

GPT-Image-2 seamless leather wallet concept with no stitching clasp or visible opening


World Knowledge and Photorealism

Shared by @levelsio

Here is a GPT-Image-2 prompt for the "World Knowledge and Photorealism" example.

Create a four-image capability showcase:
1. A full-body human anatomy poster titled "Human Anatomy - Anterior View", split between muscles, bones, organs, arteries, veins, and nerves, with dense readable medical labels and a legend.
2. A detailed physical world map poster with oceans, countries, terrain relief, compass rose, scale, and small data tables along the bottom.
3. A realistic photo of a closed Bath & Body Works storefront seen through glass doors, with a printed sign saying the store is closed for the evening.
4. A dark-mode YouTube homepage screenshot with realistic video thumbnails, sidebar navigation, channel names, view counts, and readable UI text.
Make every label and interface element sharp, aligned, and believable.

GPT-Image-2 sample image shared by levelsio GPT-Image-2 sample image shared by levelsio GPT-Image-2 sample image shared by levelsio GPT-Image-2 sample image shared by levelsio


Handwritten Medical Notes

Shared by @mark_k

Here is a GPT-Image-2 prompt for the "Handwritten Medical Notes" example.

Create a realistic photo of a doctor's handwritten clinical note on lined spiral notebook paper. Use blue ballpoint pen handwriting with abbreviated sections such as HPI, PE, A/P, vitals, dates, and medication notes. The page should look casually written during a real medical visit, with uneven spacing, natural ink pressure, and a signature at the bottom.

GPT-Image-2 text rendering sample with readable layout elements


Visual Reasoning and Structured Thinking

Shared by @daniel_mac8

Here is a GPT-Image-2 prompt for the "Visual Reasoning and Structured Thinking" example.

Create a scanned school worksheet titled "Distance and Displacement" for a student named John Doe. Include printed physics questions, a baseball diamond diagram, a running silhouette, tables for distance and displacement, multiple-choice answers, and handwritten student responses in pencil. Make the worksheet look like a real completed classroom assignment with accurate layout and readable writing.

GPT-Image-2 abstract reasoning example rendered as a visual concept


Comic Layout With Accurate Text

Shared by @itnavi2022

Here is a GPT-Image-2 prompt for the "Comic Layout With Accurate Text" example.

Draw a one-page manga with 8 panels. The main character is a chibi girl named Lisa with twin tails and a black gothic dress. She is explaining the new image generation model GPT-Image-2. Every speech bubble should contain accurate Japanese characters and readable dialogue. Keep the page clean, expressive, cute, and easy to follow.

GPT-Image-2 manga page with readable Japanese text about GPT-Image-2


Labeled Knowledge and Educational Graphics

Shared by @TrungTPhan

Here is a GPT-Image-2 prompt for the "Labeled Knowledge and Educational Graphics" example.

Create a four-image capability test for GPT-Image-2:
1. A Grand Theft Auto-style rainy nighttime scene in Hong Kong with neon Chinese signage, wet streets, a black sports car, minimap UI, subtitles, button prompts, and readable location text.
2. A Minecraft-style raid scene at night with TNT explosions, armored players, HUD hearts, inventory bar, chat messages, faction statistics, and a raid timer.
3. A typography accuracy test that places alphabet letters correctly in sequence.
Each panel should be visually distinct, highly legible, and compositionally polished.

GPT-Image-2 labeled GTA-style Hong Kong capability test GPT-Image-2 Minecraft-style space raid capability test GPT-Image-2 alphabet letter accuracy capability test


Game Concept Art and Genre Fusion

Shared by @nicdunz

Here is a GPT-Image-2 prompt for the "Game Concept Art and Genre Fusion" example.

Create a videogame key art image titled "Doom in the Backrooms." Blend the oppressive fluorescent liminal-horror look of endless yellow office corridors with the heavy-action energy of a classic demon shooter. Cinematic perspective, eerie empty space, aggressive mood, realistic lighting, game-cover composition.

GPT-Image-2 Doom in the Backrooms game-style key art


Model Upgrade Comparison

Shared by @AngryTomtweets

The original tweet compares GPT-Image-1.5 and GPT-Image-2 side by side to show how much image quality improved in one generation.

GPT-Image-1.5 versus GPT-Image-2 comparison image


Photorealistic Social and Travel Photos

Shared by @Hesamation

Here is a GPT-Image-2 prompt for the "Photorealistic Social and Travel Photos" example.

Create two realistic travel photos:
1. A sunny beach selfie of three young friends in swimwear, smiling close to the camera, with ocean waves, sand, cliffs, palm trees, sunglasses, and harsh midday sunlight.
2. A moody travel portrait of a young woman walking beside a wire fence on green coastal cliffs, with cloudy skies, distant houses, ocean water, and a natural Irish countryside atmosphere.
Make both images feel like authentic phone or travel-camera photos with believable skin, lighting, and background depth.

GPT-Image-2 realistic portrait example with natural human detail GPT-Image-2 realistic human image showing detailed portrait rendering


Chinese Text Rendering

Shared by @xiaohu

The tweet highlights GPT-Image-2's Chinese text rendering quality.

GPT-Image-2 example showing strong Chinese text rendering


Interface and Infographic Clarity

Shared by @super_bonochin

Here is a GPT-Image-2 prompt for the "Interface and Infographic Clarity" example.

Design a Japanese educational infographic titled "ChatGPTで画像を作るコツ" for an OpenAI Academy-style slide. Use four numbered columns with colorful headers, icons, checkmarks, Japanese bullet points, prompt examples along the bottom, and a simple workflow diagram. Keep the layout clean, readable, and presentation-ready.

GPT-Image-2 infographic example with readable labels and charts GPT-Image-2 information graphic example with strong text clarity


Side-by-Side Model Comparison

Shared by @itnavi2022

The tweet compares a GPT-Image-2 candidate model against NanoBanana 2 side by side.

GPT-Image-2 side-by-side comparison sample against another image model GPT-Image-2 model comparison image showing visual quality differences


Image Quality at Zoom Level

Shared by @mark_k

The tweet is useful because it shows a real limitation: GPT-Image-2 can look great at normal viewing size but still break down in fine details when you zoom in.

GPT-Image-2 example that looks strong at normal viewing distance GPT-Image-2 zoomed detail example showing artifact issues


Cinematic Driving Close-Up

Shared by @synthwavedd

Here is a GPT-Image-2 prompt for the "Cinematic Driving Close-Up" example.

Create a cinematic close-up shot of a bald man driving a car, seen through the windshield. He has a serious expression, one hand on the steering wheel, warm sunlight and shadow bands across his face, realistic car interior fabric, shallow depth of field, and a gritty film still look.

GPT-Image-2 example containing an unexpected generated watermark artifact


Anime and Character Scene Rendering

Shared by @studio_oneroom

Here is a GPT-Image-2 prompt for the "Anime and Character Scene Rendering" example.

Create an anime-style key visual with a detailed character in the foreground, a rich illustrated background, polished lighting, readable embedded text, and strong harmony between character art and environment design. High detail, clean linework, vibrant but balanced colors, polished poster-like finish.

GPT-Image-2 anime-style scene with strong text and character rendering


Realistic Travel and Vacation Photos

Shared by @PurzBeats

Here is a GPT-Image-2 prompt for the "Realistic Travel and Vacation Photos" example.

Create a realistic close-up vacation selfie of a heavyset middle-aged man with short gray hair and a gray beard, making an excited wide-eyed expression from a hotel balcony in Hawaii. Show palm trees, high-rise buildings, blue sky, and bright daylight in the background. Make it feel like an authentic phone selfie.

GPT-Image-2 realistic vacation-style image with convincing travel photography look


Minecraft-Style Brand Environment

Shared by @BLCNYY

Here is a GPT-Image-2 prompt for the "Minecraft-Style Brand Environment" example.

Create a Minecraft-style first-person screenshot inside "Claude Headquarters." Show blocky office architecture, the Claude logo on the wall, a desk with a confidential document, and a held paper titled "Internal Document: Claude Opus 5" listing next-generation model features. Include Minecraft HUD elements, inventory bar, chat text, pixelated lighting, and readable document typography.

GPT-Image-2 single-image realism showcase


Character Makeover and Meme Images

Shared by @marmaduke091

Here is a GPT-Image-2 prompt for the "Character Makeover and Meme Images" example.

Create an anime fantasy character poster of MrBeast reimagined as a dark armored demon warrior. Show a confident male hero with a recognizable face, black-and-blue spiked armor, exposed torso, dramatic low-angle pose, glowing blue demonic creature behind him, vivid magenta and blue lighting, and readable "MR BEAST" text on a banner in the background.

GPT-Image-2 makeover-style meme image example


Branded Satire and Corporate Mashups

Shared by @stark_nico99

Here is a GPT-Image-2 prompt for the "Branded Satire and Corporate Mashups" example.

YouMind 收购了 ChatGPT 的图

GPT-Image-2 fictional corporate acquisition satire image


Same Prompt, Better Realism

Shared by @phasE89

Here is a GPT-Image-2 prompt for the "Same Prompt, Better Realism" example.

Amatérská fotka staršího páru sedícího v yorkshirské hospodě, amatérská kompozice, momentka.

Older public image model result from the same prompt GPT-Image-2 preview result from the same prompt with stronger realism


Mixed Style Text and Character Tests

Shared by @itnavi2022

Here is a GPT-Image-2 prompt for the "Mixed Style Text and Character Tests" example.

Create a four-image test set for text, realism, and character rendering:
1. A realistic newspaper on a cafe table with the headline "OpenAI Unveils GPT-Image-2", columns of readable article text, coffee, glasses, water, pastry, and a pen.
2. A Japanese TV weather broadcast screenshot from Shibuya at 6:45, with live-news overlays, weather icons, temperature, humidity, wind speed, and readable Japanese captions.
3. A photorealistic cat-ear maid cosplayer eating soft-serve ice cream on the Spanish Steps in Rome, with tourists and architecture in the background.
4. A clean anime illustration of a cheerful twin-tail gothic girl waving, with a speech bubble saying "Hi!".

GPT-Image-2 Japanese illustration sample one GPT-Image-2 Japanese illustration sample two GPT-Image-2 Japanese illustration sample three GPT-Image-2 Japanese illustration sample four


GPT-Image-1.5 Versus GPT-Image-2 in Illustration Quality

Shared by @itnavi2022

The original tweet compares two outputs and asks viewers to identify which one was generated by GPT-Image-2.

Image model comparison sample one in anime illustration style Image model comparison sample two in anime illustration style


Prompting for Complex Religious Text on Objects

Shared by @TlanoAI

Original idea from the tweet:

Here is a GPT-Image-2 prompt for the "Prompting for Complex Religious Text on Objects" example.

般若心経が書かれたバナナ

GPT-Image-2 banana with dense Japanese text as a complex text-rendering challenge


Comparison Against Nano Banana Pro

Shared by @thefinnmckenty

The tweet compares GPT-Image-2 against Nano Banana Pro using the same all-American stunt-show idea, with a motorcycle jump, explosions, arena lighting, and American flag screens.

GPT-Image-2 versus Nano Banana Pro comparison image one GPT-Image-2 versus Nano Banana Pro comparison image two


This list is updated regularly. If you have a prompt you want added, share it on Twitter and tag it with GPT-Image-2.


Sci-Fi Movie Poster With Japanese Typography

Shared by @studio_oneroom

Here is a GPT-Image-2 prompt for the "Sci-Fi Movie Poster With Japanese Typography" example.

画像生成には必ず最新のモデル GPT-image-2を使ってください

添付1の背景を添付2の画像に変えて その際にうまく馴染むように構図やライティングを整えて映画のポスター風にしてください

タイトル:アルゴディアス
キャプション: 2046年、ドル崩壊でアメリカが覇権を失い、世界は欧州防衛機構(EDA)と南半球資源同盟(SRA)に二分された。太平洋の制海権を巡る戦火はニュージーランドに達し、ウェリントン基地がEDA最前線となる。戦況が膠着する中、中立を掲げる技術大国・日本が戦局を覆す新型機動兵器「アルゴディアス」を極秘開発。その兵器が間もなくウェリントンに配備されるという...

GPT-Image-2 sci-fi movie poster with a towering mech and Japanese title text


Everyday Lifestyle Realism

Shared by @WolfRiccardo

Here is a GPT-Image-2 prompt for the "Everyday Lifestyle Realism" example.

1. A candid apartment doorway photo of a young woman in a simple black dress holding a small decorated birthday cake, shot with direct flash, realistic skin texture, casual home lighting, slightly messy interior, authentic smartphone photo look.

2. A warm indoor lifestyle portrait of a young woman sitting on a sofa in a black top and white skirt, looking at her phone, soft evening light, natural pose, cozy apartment background, photorealistic smartphone snapshot.

3. A live concert crowd photo of a female singer on stage under purple and pink lights, audience hands raised in the foreground, realistic arena lighting, motion and depth, captured like an iPhone concert photo.

4. A bright luxury poolside vacation photo of a woman in sunglasses relaxing beside a turquoise swimming pool, summer resort atmosphere, striped towel, palm shadows, glossy travel photography style.

GPT-Image-2 realistic doorway birthday snapshot GPT-Image-2 realistic sofa lifestyle portrait GPT-Image-2 realistic concert crowd phone photo GPT-Image-2 realistic luxury poolside vacation photo


POV and First-Person Angles

Shared by @WolfRiccardo

Here is a GPT-Image-2 prompt for the "POV and First-Person Angles" example.

1. First-person view from the driver's seat of a Tesla at a red light in a modern city, hands on the steering wheel, dashboard screen visible, realistic windshield reflections, natural daylight, smartphone photo realism.

2. Police bodycam-style night scene during a traffic stop, first-person perspective, one officer in frame, headlights and streetlights reflecting on wet pavement, gritty realism, documentary look.

3. First-person mountain biking shot on a narrow dirt trail through a green forest, handlebars visible, dynamic motion blur, sunlight through trees, action camera realism.

4. First-person view at a convenience store checkout counter, cashier and customers looking toward the camera, snack shelves and refrigerators in the background, fluorescent store lighting, believable candid photo style.

GPT-Image-2 first-person Tesla interior photo GPT-Image-2 first-person police bodycam traffic stop GPT-Image-2 first-person mountain biking action scene GPT-Image-2 first-person convenience store scene


Human Anatomy Teaching Poster

Shared by @drkeithsiau

Here is a GPT-Image-2 prompt for the "Human Anatomy Teaching Poster" example.

Create a clean medical teaching poster showing the anterior view of human anatomy. Include readable labels for major muscles, bones, organs, arteries, and veins. Use a neutral background, textbook layout, accurate proportions, and crisp educational diagram styling.

GPT-Image-2 medical anatomy poster with detailed labels and anterior body view


Anime Mecha Concept Sheet

Shared by @old_pgmrs_will

Here is a GPT-Image-2 prompt for the "Anime Mecha Concept Sheet" example.

Create a concept sheet for a next-generation anime protagonist mecha. Show a towering black-and-silver robot with glowing accents, multiple technical callouts, front and side details, weapon notes, small pilot references, and polished industrial design presentation on a dark studio background.

GPT-Image-2 anime mecha concept sheet with technical callouts


Rough Colored-Pencil Plant Study

Shared by @Masimo_Blue

Here is a GPT-Image-2 prompt for the "Rough Colored-Pencil Plant Study" example.

Create rough colored-pencil concept sketches of a small potted seedling with simple gardening hardware nearby. Keep the lines uneven, the coloring loose, and the overall drawing slightly amateur and natural rather than polished or perfect.

GPT-Image-2 rough colored-pencil sketch of a potted seedling GPT-Image-2 rough colored-pencil gardening concept sketch


Jewel-Toned Mermaid Fantasy

Shared by @NeneneAI

Here is a GPT-Image-2 prompt for the "Jewel-Toned Mermaid Fantasy" example.

Create a luminous fantasy illustration of a pink-haired mermaid princess swimming through an underwater palace filled with glowing crystals, pearls, gold ornaments, and floating bubbles. Use rich jewel tones, sparkling water caustics, elegant dress-like fins, and a magical storybook atmosphere.

GPT-Image-2 jewel-toned mermaid fantasy illustration with underwater palace details


Japanese Magazine Cover Layout

Shared by @old_pgmrs_will

Here is a GPT-Image-2 prompt for the "Japanese Magazine Cover Layout" example.

Design a bold Japanese magazine cover packed with readable headlines, subheads, badges, and editorial callouts. Center the layout around a festive New Year special theme with a confident host portrait, bright red-and-gold accents, and dense, realistic print-magazine typography.

GPT-Image-2 Japanese magazine cover with dense editorial typography


Anime Suit Portrait

Shared by @mirochill

Here is a GPT-Image-2 prompt for the "Anime Suit Portrait" example.

Create a stylish anime portrait of a young man in a charcoal suit and tie, leaning back in a chair with one leg crossed. Use a low camera angle, soft office lighting, clean modern background, polished hair, and a calm confident expression.

GPT-Image-2 anime portrait of a young man in a suit from a low angle


Floating Fantasy City at Dusk

Shared by @old_pgmrs_will

Here is a GPT-Image-2 prompt for the "Floating Fantasy City at Dusk" example.

Create an epic fantasy city floating high above the clouds at dusk. Show glowing bridges, layered towers, hanging gardens, waterfalls dropping into the sky, and warm sunset light breaking through dramatic clouds. Make it feel like a grand matte painting for a fantasy film.

GPT-Image-2 floating fantasy city above the clouds at dusk


Japanese Infographic Before-and-After

Shared by @takuto_pitaliy

The tweet compares an older result with an updated GPT-Image-2-style result using the same prompt.

Here is a GPT-Image-2 prompt for the "Japanese Infographic Before-and-After" example.

Create a Japanese infographic poster that explains how to identify fever symptoms in children. Use a cute character illustration, a warm clinic-friendly color palette, clear section blocks, temperature callouts, and dense but readable Japanese text.

Older infographic-style result with weaker Japanese text rendering GPT-Image-2-style Japanese infographic with cleaner layout and typography


Realistic Engineer's Screen

Shared by @CuriousRefuge

Here is a GPT-Image-2 prompt for the "Realistic Engineer's Screen" example.

average engineer's screen

GPT-Image-2 realistic cluttered engineer workstation screen and desk setup


English and Japanese UI Rendering Test

Shared by @itnavi2022

Here is a GPT-Image-2 prompt for the "English and Japanese UI Rendering Test" example.

Create two side-by-side web page mockups for a text rendering test. One version should use English UI text and the other should use Japanese UI text. Keep the layouts similar, make the typography readable, and present both like polished website screenshots.

GPT-Image-2 English website text rendering test screenshot GPT-Image-2 Japanese website text rendering test screenshot


Battle Royale Lobby Screenshot

Shared by @SuguruKun_ai

Here is a GPT-Image-2 prompt for the "Battle Royale Lobby Screenshot" example.

Create a realistic battle royale lobby screenshot with four recognizable tech leaders standing side by side as playable characters. Include a polished game interface, party status panels, buttons, rank badges, and bright background lighting that feels like a live multiplayer game menu.

GPT-Image-2 battle royale lobby screenshot with recognizable tech leader avatars


Weapon Skin Mockup in First Person

Shared by @WolfRiccardo

Here is a GPT-Image-2 prompt for the "Weapon Skin Mockup in First Person" example.

Create a first-person shooter weapon skin mockup shown from the player's hands. The rifle should have a playful pink-and-blue custom finish with sticker details, polished highlights, and a clean game-inventory presentation style.

GPT-Image-2 first-person weapon skin mockup with bright custom finish


Real-World Documents and Packaging

Shared by @mark_k

Here is a GPT-Image-2 prompt for the "Real-World Documents and Packaging" example.

1. Create a realistic printed medical intake form filled out by hand on a clipboard, with believable signatures, checkboxes, notes, and pen pressure variations. Keep the paper texture and lighting natural like a phone snapshot.

2. Create a realistic geological borehole log sheet with soil layers, handwritten field notes, measurements, chart markings, and engineering annotations on worn paper.

3. Design a cereal box package for "Corn Flakes" with a believable supermarket-brand layout, nutrition panel, cheerful mascot, and crisp readable packaging text.

4. Create a casual smartphone selfie of Barack Obama and Donald Trump together indoors, smiling toward the camera, with realistic skin, lighting, and social-media snapshot framing.

GPT-Image-2 realistic handwritten medical form on a clipboard GPT-Image-2 realistic geological borehole log sheet GPT-Image-2 cereal box packaging with readable branding and label text GPT-Image-2 selfie of Barack Obama and Donald Trump taken like a casual phone photo


Manga-Style Japanese Explainer Panel

Shared by @itnavi2022

Here is a GPT-Image-2 prompt for the "Manga-Style Japanese Explainer Panel" example.

Create a manga-style Japanese explainer panel about testing a new AI image model. Use bold headline text, speech bubbles, side notes, and a smartphone screen insert. The layout should feel like a social media tutorial image with readable Japanese typography.

GPT-Image-2 manga-style Japanese explainer panel with readable labels and phone UI


Product Launch Keynote Satire

Shared by @yousefrol

Here is a GPT-Image-2 prompt for the "Product Launch Keynote Satire" example.

Create a satirical Apple-style keynote slide introducing a strange new device. Show a presenter on stage, a giant product render behind them, dramatic presentation lighting, minimalist branding, and clean readable event text like a real launch announcement.

GPT-Image-2 satirical keynote presentation slide with a bizarre product reveal


Composite Scene From Multiple References

Shared by @studio_oneroom

Here is a GPT-Image-2 prompt for the "Composite Scene From Multiple References" example.

添付1の背景を添付2の画像に変えて その際にうまく馴染むように構図やライティングを整えて自然な一枚絵にしてください

The tweet shows the generated image alongside the two source references.

GPT-Image-2 composite scene blending a subject into a new background naturally Source reference image one from the tweet Source reference image two from the tweet


GTA-Style Cinematic Game Screenshots

Shared by @marmaduke091

Here is a GPT-Image-2 prompt for the "GTA-Style Cinematic Game Screenshots" example.

Create cinematic screenshots from a fictional open-world crime game set in a bright coastal city. Show realistic characters, beachside streets, driving scenes, dialogue subtitles, warm sunset light, and a polished in-game HUD that feels like a modern AAA console game.

GPT-Image-2 GTA-style cinematic dialogue screenshot with two characters GPT-Image-2 GTA-style open-world driving screenshot on a sunny street


Lucia and Jason Game Screenshot

Shared by @GTAInfinity_YT

Here is a GPT-Image-2 prompt for the "Lucia and Jason Game Screenshot" example.

a screenshot from gta 6 with lucia and jason

The thread also includes a no-HUD version of the same idea.

GPT-Image-2 GTA 6 style screenshot with Lucia and Jason by the water GPT-Image-2 no-HUD GTA 6 style scene with Lucia and Jason at sunset


Jason Walking in Leonida Keys

Shared by @GTAInfinity_YT

Here is a GPT-Image-2 prompt for the "Jason Walking in Leonida Keys" example.

a screenshot from gta 6 with Jason walking in leonida keys

GPT-Image-2 GTA 6 style screenshot of Jason walking through Leonida Keys


Mechanical Keyboard Design Grid

Shared by @TanShilong

Here is a GPT-Image-2 prompt for the "Mechanical Keyboard Design Grid" example.

一个九宫格图片,展现九位当代知名设计师设计的同一组物体:机械键盘,包括设计师头像,设计师对于设计的中文文字解读和作品呈现。排版统一规则

GPT-Image-2 nine-panel mechanical keyboard design grid with Chinese annotations


Basketball Court Flash Portrait

Shared by @BubbleBrain

Here is a GPT-Image-2 prompt for the "Basketball Court Flash Portrait" example.

35mm color film photography with harsh direct on-camera flash, specular highlights on skin and clothing, strong catchlights in eyes, high contrast flash illumination, authentic film grain and color shift, high fashion fresh innocent basketball court editorial style, intimate first-person low-angle POV shot from below, early 20s sexy Chinese female idol with ultra-realistic delicate refined Chinese features, seductive almond-shaped fox eyes with natural double eyelids, high nose bridge, small sharp V-shaped jawline, flawless realistic porcelain skin with cool ivory undertone and visible flash specular highlights, fine delicate skin texture with subtle pores micro details and natural dewy glow under flash, fresh natural sporty makeup with soft dewy glow, subtle natural flush on cheeks, natural pink lips slightly parted, subtle natural freckles across nose and cheeks, long dark brown hair tied in a high playful ponytail with some loose strands framing the face and realistic loose strands, wearing a loose white tank top and white high-waisted basketball shorts, white knee-high sports socks, seductive natural leaning pose against the basketball hoop pole on the outdoor court at dusk, body angled sideways with naturally arched back and hips gently pushed back to accentuate perky round hips and sexy butt curve, one leg naturally extended forward toward the camera and the other leg slightly bent to emphasize long sexy legs, both hands lightly resting on the basketball pole at shoulder height, intensely seductive playful yet pitiable doe-eyed gaze straight at the viewer with soft vulnerable longing eyes and a gentle teasing smile full of quiet temptation and desire, harsh direct on-camera flash creating sharp specular highlights and strong catchlights, background with blurred basketball court and hoop under dusk sky, high contrast film color grading with natural flash look, extremely sharp yet soft skin rendering with authentic 35mm direct flash aesthetic, natural hair strands, realistic fabric texture on tank top and shorts with socks detail, no plastic skin, no digital over-sharpening, no airbrushing, no blemishes, no moles, no oily skin, no watermark, no text, authentic 35mm direct flash film basketball court look --ar 9:16

GPT-Image-2 35mm flash portrait on an outdoor basketball court at dusk


Old Game Character Sheet

Shared by @os_orabi

Here is a GPT-Image-2 prompt for the "Old Game Character Sheet" example.

Turn this single reference image of an old game character into a complete production-ready character sheet. Include front, side, and back views, facial expressions, close-up head details, accessories, material notes, weapon references, and a clean layout suitable for 3D modeling or concept art.

The tweet shows the generated character sheet with the original game reference.

GPT-Image-2 full character sheet generated from an old game character reference Source reference image for the old game character sheet


Catalog Outfit Street Style

Shared by @VORTEX_Promos

Here is a GPT-Image-2 prompt for the "Catalog Outfit Street Style" example.

street style shot of a woman in this outfit.

GPT-Image-2 street style outfit generation from catalog clothing references


Onsen Ryokan Film Portrait

Shared by @BubbleBrain

Here is a GPT-Image-2 prompt for the "Onsen Ryokan Film Portrait" example.

35mm film photography, warm vintage Japanese onsen ryokan aesthetic, soft ambient wooden lantern lighting mixed with gentle natural window light, subtle film grain, gentle color shift, high atmosphere editorial style, intimate medium shot, early 20s beautiful Chinese female idol with ultra-realistic delicate refined Chinese features, seductive almond-shaped fox eyes with natural double eyelids, high nose bridge, small sharp V-shaped jawline, flawless porcelain skin with warm ivory undertone, visible subtle skin texture and micro pores, soft natural makeup with dewy glow, subtle rosy flush on cheeks, natural soft pink lips slightly parted, long dark brown hair tied in a loose low bun with some messy strands falling around face and neck, wearing a loose white yukata (traditional Japanese bathrobe) deliberately slipped off one shoulder and loosely tied at the waist, the fabric slightly open revealing smooth skin and subtle cleavage, barefoot, seductive relaxed sitting pose on the edge of a traditional wooden engawa veranda at a vintage onsen ryokan, body slightly turned toward the camera, one leg bent with foot resting on the wooden floor, the other leg gently dangling, one hand lightly holding the yukata collar, the other hand resting on the wooden floor behind her for support, softly arched back to gently accentuate curves, intensely seductive yet gentle and inviting gaze straight at the viewer with soft doe eyes full of quiet temptation and warmth, warm wooden interior with paper sliding doors and distant steaming hot spring in soft focus, gentle rim lighting highlighting skin and fabric texture, authentic vintage film color grading with warm tones, extremely sharp yet soft skin rendering, natural hair strands, realistic fabric wrinkles and drape on the yukata, no plastic skin, no digital over-sharpening, no airbrushing, no blemishes, no moles, no oily skin, no watermark, no text, authentic 35mm film Japanese onsen ryokan atmosphere

GPT-Image-2 warm 35mm film portrait in a vintage Japanese onsen ryokan


SaaS Homepage and Pricing Flow

Shared by @qiufenghyf

Here is a GPT-Image-2 prompt for the "SaaS Homepage and Pricing Flow" example.

modern SaaS website homepage, figma style, clean navigation bar, hero section with floating prompt input UI, rounded card, soft shadow, minimal design

background filled with colorful creative design boards, portfolio layouts, typography posters, UI mockups, grid composition

centered composition, elegant spacing, premium UI design, soft lighting, subtle depth, product landing page, 8k, ultra clean

The tweet shows the model continuing from a homepage into pricing and feature pages.

GPT-Image-2 SaaS homepage mockup with prompt input hero section GPT-Image-2 SaaS pricing page mockup continued from the homepage design GPT-Image-2 SaaS feature page mockup with consistent visual design


Late-Night Convenience Store Portrait

Shared by @BubbleBrain

Here is a GPT-Image-2 prompt for the "Late-Night Convenience Store Portrait" example.

35mm film photography with harsh convenience store fluorescent lighting mixed with colorful neon signs from outside, authentic film grain, high contrast, slight color cast, cinematic street editorial style, intimate medium shot, early 20s sexy Chinese female idol with ultra-realistic delicate refined Chinese features, seductive almond-shaped fox eyes with natural double eyelids, high nose bridge, small sharp V-shaped jawline, flawless porcelain skin with cool ivory undertone and visible specular highlights from fluorescent light, subtle skin texture and micro pores, natural dewy makeup with soft flush on cheeks, glossy natural pink lips slightly parted, subtle natural freckles across nose and cheeks, long dark brown hair in a messy high ponytail with many loose strands falling around face and neck, wearing an oversized white button-up shirt as the only top, unbuttoned at the top with deep cleavage and loosely tied at the waist, paired with a tiny black pleated mini skirt, barefoot in simple white slides, seductive casual leaning pose against the glass door of a 24-hour convenience store at late night, body slightly arched, one leg bent with foot resting against the door frame, the other leg straight, one hand holding a bottle of iced drink, the other hand lightly pulling the hem of her mini skirt, intensely seductive playful yet slightly vulnerable gaze straight at the viewer with soft doe eyes full of quiet temptation and teasing smile, bright cold fluorescent store light from inside mixed with pink and blue neon glow from outside signs, realistic reflections on glass door, blurred convenience store interior with shelves and snacks in background, authentic 35mm film color grading with harsh lighting and neon accents, extremely sharp yet soft skin rendering, natural hair strands, realistic fabric wrinkles and drape on the oversized shirt and mini skirt, no plastic skin, no digital over-sharpening, no airbrushing, no blemishes, no moles, no oily skin, no watermark, no text, authentic late-night convenience store atmosphere

GPT-Image-2 35mm film portrait at a late-night convenience store


White Studio Tennis Portrait

Shared by @BubbleBrain

Here is a GPT-Image-2 prompt for the "White Studio Tennis Portrait" example.

Professional studio fashion photography, ultra-clean high-end digital editorial portrait in a pure white seamless cyclorama photography studio, massive soft octabox key light + large softbox fill light creating flawless even illumination with gentle sculpting shadows, Vogue-level minimalist high fashion aesthetic, eye-level shot, 3/4 body shot, early 20s Korean idol-type young woman with ultra-realistic delicate refined Korean features, slightly upturned almond-shaped fox eyes with natural double eyelids, small sharp jawline, flawless yet realistic porcelain skin with cool ivory undertone, completely natural post-workout no-makeup look, soft natural flush on cheeks, subtle dewy skin glow from light sweat, natural dusty rose lips slightly parted, subtle natural freckles across nose and cheeks, visible subtle skin texture, pores and micro skin details, long naturally wavy dark brown hair slightly messy with realistic loose strands, wearing a white fitted tennis polo or short-sleeve athletic top and a high-waisted white tennis skirt, standing naturally in the center of the pure white studio, holding a white tennis racket casually resting on one shoulder, the other hand relaxed at her side or lightly on her hip, one leg slightly forward, posture natural and confident, looking directly at the viewer at eye level with a calm, confident and fresh expression, pure soft professional studio lighting with large diffused sources, flawless even illumination, gentle sculpting shadows that naturally define the figure, clean high-key minimalist aesthetic, ultra-clean high-end digital photography, extremely sharp yet soft skin rendering with natural dewy glow, no film grain, no noise, no halation, no vignette, perfect neutral color grading, authentic Korean idol studio editorial aesthetic, visible realistic tennis racket texture, natural skin texture with freckles and subtle sweat sheen, no plastic skin, no digital over-sharpening, no airbrushing, no watermark, no text --ar 9:16

GPT-Image-2 high-key white studio tennis portrait with racket


Spring 2026 Boston City Poster

Shared by @BubbleBrain

Here is a GPT-Image-2 prompt for the "Spring 2026 Boston City Poster" example.

A striking Spring 2026 city poster for Boston with an elegant celebratory mood and a bold contemporary design. On a clean off-white textured background with large areas of negative space, a miniature single sculler rows across the lower right corner of the image on a narrow ribbon of reflective water. The wake from the oar sweeps upward in a dynamic calligraphic curve, gradually transforming into the Charles River and then into a dreamlike hand-painted panorama of Boston. Inside this flowing river-shaped composition are iconic Boston elements: the Back Bay skyline, Beacon Hill brownstones, Acorn Street, Boston Public Garden, Swan Boats, Zakim Bridge, Fenway-inspired details, historic brick architecture, harbor ferries, and the city’s waterfront atmosphere. Soft morning fog, golden spring light, subtle festive accents in crimson and gold, rich detail, layered depth, sophisticated city-poster aesthetics, fresh and refined, visually powerful but not overcrowded. Elegant typography in the lower left reads “SPRING 2026” with a vertical slogan “BOSTON, A CITY OF RIVER, MEMORY, AND INVENTION”, text clear and beautifully composed, premium graphic design, 9:16

GPT-Image-2 Spring 2026 Boston city poster with rower and river-shaped skyline


Shituoling Fantasy Scene Comparison

Shared by @MANISH1027512

This tweet compares MJ v8.1 and GPT Image 2 on the same fantasy scene prompt. The original post does not label which image is which, so the images are kept as a side-by-side comparison without assigning model names to the order.

中式怪异,黑暗神秘风格融合中式美学,完美细节,多重管线渲染,完美建模。西游记背景,狮驼岭,千妖万怪,坐在左边巨大王座上的大象王重甲妖精,坐在中间巨大王座上的狮王重甲妖精,坐在右边巨大王座上大鹏鸟王重甲妖精。渺小的背对镜头孙悟空肩抗金箍棒步行前进,孙悟空身穿铠甲,近地仰拍镜头,长焦镜头,强烈阴影。极致细节刻画,多次修改,正确透视和主体线条,精致细节

MJ v8.1 and GPT Image 2 comparison output for a dark Chinese fantasy Shituoling scene MJ v8.1 and GPT Image 2 comparison output with Sun Wukong facing three demon kings


Reference Pose Test Against Nano Banana 2

Shared by @desds1678

This tweet compares the same prompt and reference image in GPT Image 2 and Nano Banana 2. The author labels the left image as GPT Image 2 and the right image as Nano Banana 2.

Here is a recreated prompt for the visible comparison.

Use the reference image to create a realistic full-body fashion portrait of a young woman sitting indoors on a white sofa near soft window light. Keep the same general pose, body proportions, casual white top, gray skirt, black tights, natural room lighting, and simple clean background. Make the result look like a realistic lifestyle photo.

GPT Image 2 comparison output for a reference pose indoor fashion portrait Nano Banana 2 comparison output for the same reference pose indoor fashion portrait