The Image Generation Model That's Hotter Than Nano Banana Has Leaked, Screenshots Are No Longer Evidence | Includes Prompts

marsbitPublicado em 2026-04-19Última atualização em 2026-04-19

Resumo

A new AI image generation model, widely referred to as "GPT Image 2," has been leaked and is demonstrating significant advancements over predecessors like DALL-E 3 and even Google's Nano Banana Pro. It excels in four key areas: text rendering, prompt adherence, photorealism, and world knowledge. The model can generate highly accurate text in multiple languages, including complex Chinese characters, making it capable of producing convincing fake documents, UI screenshots, and product labels. This capability also raises concerns about the reliability of using screenshots as evidence. The model is currently in A/B testing, with a full release expected around May 2026 when DALL-E services are officially retired. It is accessible for testing on the LM Arena platform. The article includes several prompt templates optimized for the model, such as generating realistic app screenshots, product photos with detailed labels, and street scenes with accurate signage. This advancement is reshaping creative workflows but also accelerating the displacement of some traditional design roles.

Is your impression of text-to-image still stuck on Nano Banana?

But kid, times have changed again.

@johnAGI168 https://x.com/johnAGI168/status/2044781168151724067

@0115hippo https://x.com/0115hippo/status/2044722124611539160

In early April, three anonymous image models, codenamed maskingtape-alpha, packingtape-alpha, and gaffertape-alpha, appeared on the LM Arena evaluation platform. They disappeared a few hours later.

OpenAI has not officially announced this model yet, but based on the metadata returned by the API and user-side testing records, it has already gained a widely accepted name: GPT Image 2.

Screenshots Can No Longer Be Used as Evidence

Over the past few years, one of the most obvious weaknesses of AI image generation models has been text within images. In the DALL-E 3 era, if you asked it to write "Hello" in an image, it might output "Hellp" or even "Hl10", with letters tilting drunkenly. GPT Image 1 improved a lot, handling simple English labels. By GPT Image 1.5, its accuracy in rendering English text was close to 95%, but it still had significant flaws with non-Latin scripts like Chinese, Japanese, and Korean.

But the leaked sample images from GPT Image 2 have changed this impression.

@MrLarus https://x.com/MrLarus/status/2044824800909054181

@akokoi1 https://x.com/akokoi1/status/2044789531615056175

The text in the images is exactly what it should be. Chinese characters are clear, with accurate glyphs and complete strokes. Someone tested generating an ID card-style image, where the name, address, and ID number were all rendered correctly, with neat formatting, looking at first glance like a photo of a real document.

This is good news. The improvement in text rendering means generating infographics, posters, product packaging, and complex charts becomes more reliable.

But there's always another side to the coin. A model that can generate photo-realistic ID-style images and precisely render UI screenshots naturally makes "screenshots can be used as evidence" increasingly questionable.

By comparison, this is also a core difference between the GPT Image series and other models. Midjourney still has no progress in text rendering, and the Stable Diffusion series also has this old problem. According to the leaked Arena test results, GPT Image 2 surpassed Midjourney in four dimensions: text rendering, instruction following, photorealism, and world knowledge. Midjourney's advantages are mainly retained in artistic style and aesthetic control.

Does It Really Know What the World Looks Like?

A tester asked the model to generate a hypothetical GPT-8 product pricing page. The resulting image had a layout that was indeed in the style of the OpenAI website, with button placement and font choices resembling those from a real interface, and the hierarchical logic of the price table was correct.

GPT Image 2 can generate images extremely similar to real software interfaces, including browser windows, mobile app interfaces, and data visualization charts, with a level of fidelity unmatched by the previous generation.

@johnAGI168 https://x.com/johnAGI168/status/2044781168151724067

@levelsio https://x.com/levelsio/status/2040333489476681758

This will lead to some very interesting practical uses. When designers are creating product prototypes, they don't need to open Figma first and draw a bunch of wireframes; they can directly describe the desired interface in text, and the output is a reference image that can be used for team discussions. When creating investor decks, they can show a "product screenshot" without waiting for an engineer to write code. When writing documentation, example interface images for illustration can be generated directly, without having to think about where to find screenshots for a blank page.

@marmaduke091 https://x.com/marmaduke091/status/2040338311873515597

Image Generation Is No Longer Just "Image Generation"

OpenAI has already announced that DALL-E 2 and DALL-E 3 will officially cease service on May 12, 2026. Azure OpenAI's DALL-E 3 was retired early in February.

DALL-E was the first place many people encountered AI image generation, from those blurry early works to today, in just a few short years.

Meanwhile, Google, which had just established its industry position with Nano Banana Pro in early 2026, might feel the pressure. Early test reports indicate that GPT Image 2 simultaneously surpasses Nano Banana Pro in three dimensions: realism, text rendering, and world knowledge. This kind of triple win is not common.

For creators, the feeling is complex. Illustrators, graphic designers, and photographers are not facing this topic for the first time. Since the release of GPT Image 1, the number of freelance graphic design positions has decreased by about 18%. AI has indeed replaced the decision to "hire someone to do this" in certain scenarios, but it is also creating new ways of working, allowing one person to do more.

The evolution speed of image generation models no longer leaves much time for adaptation. It was only a few months from GPT Image 1's launch to version 1.5. And from 1.5 to 2, it's only been about half a year. Each generation solves the core shortcomings of the previous one while opening up new possibilities.

GPT Image 2 is currently still in the A/B testing phase, with some ChatGPT users randomly gaining access. The official release window is widely predicted to be around May, coinciding with the retirement of DALL-E. If you want to experience it early, you can currently try your luck on the LM Arena evaluation platform.

Test Address: https://arena.ai

Based on community feedback and the known strengths of this model, the following prompt templates can maximize your chances of success:

UI/Screenshot Prompt: A photorealistic screenshot of a mobile banking app, clearly showing transaction history with dates, amounts, and merchant names legible. iPhone 16 screen, natural hand holding the phone, coffee shop background.

Product Label Prompt: A photographic product photo of a craft beer bottle, with clear label details showing the brewery name "Oakridge Brewing Co.", alcohol content 6.8%, a mountain logo, and an ingredient list. Studio lighting, white background.

Signage Prompt: A street scene photo of a Tokyo alley at night, showing multiple neon signs in both Japanese and English, including a ramen shop sign reading "Ichiban Ramen — Est. 1987", a karaoke bar sign, and various glowing advertisements. Wet, reflective pavement with light reflections.

Interface/World Knowledge Prompt: A photorealistic YouTube video screenshot showing a video titled "How to Assemble a Computer in 2026" with 2.3 million views, featuring realistic comments, sidebar video recommendations, and channel info. Desktop browser view.

Widescreen Trigger Prompt: A cinematic widescreen photo of an IKEA store exterior at dusk, showing the glowing IKEA sign, a parking lot with realistic cars, and shoppers entering and leaving. Golden hour lighting, 16:9 format.

Unattributed image sources and references: https://miraflow.ai/blog/how-to-use-duct-tape-ai-model-arena-gpt-image-2-guide

This article is from the WeChat public account "APPSO", author: Discovering Tomorrow's Products

Criptomoedas em alta

ctr
CitreaCTR
wstusdt
wrapped stUSDTWSTUSDT
apr
aPrioriAPR
ctx
Cryptex FinanceCTX
audio
AudiusAUDIO
mantra
MantraMANTRA
xvs
VenusXVS
waxl
AxelarWAXL
bill
Billions NetworkBILL
pyth
PYTH (Pyth)PYTH
rune
THORChainRUNE
velodrome
Velodrome FinanceVELODROME
brev
BrevisBREV
zrx
ZRX（0X）ZRX
cake
PancakeSwapCAKE

Perguntas relacionadas

QWhat is the name of the leaked image generation model mentioned in the article, and what is its significance?

AThe leaked model is referred to as GPT Image 2. Its significance lies in its dramatic improvement in text rendering accuracy, especially for non-Latin scripts like Chinese, and its ability to generate highly realistic images, including convincing UI screenshots and document-style images, which challenges the reliability of screenshots as evidence.

QHow does GPT Image 2's performance compare to other models like Midjourney and Google's Nano Banana Pro?

AAccording to the article, GPT Image 2 outperforms Midjourney in text rendering, prompt following, photorealism, and world knowledge, with Midjourney retaining an advantage mainly in artistic style and aesthetic control. It also reportedly surpasses Google's Nano Banana Pro in realism, text rendering, and world knowledge.

QWhat are some of the potential practical applications of GPT Image 2's capabilities?

APotential applications include generating product prototypes and UI mockups for designers, creating realistic 'screenshots' for investor decks without coding, producing example interface images for documentation, and generating accurate product labels, packaging, and information graphics.

QWhat major change is OpenAI making to its image generation services in relation to this new model?

AOpenAI has announced that DALL-E 2 and DALL-E 3 will officially stop service on May 12, 2026, with Azure's DALL-E 3 having already been retired in February. This suggests a transition to the new GPT Image model series.

QWhere can users currently try to access or test the GPT Image 2 model, and what is a recommended strategy for getting good results?

AThe model is currently in A/B testing, with some ChatGPT users randomly gaining access. Users can also try their luck on the LM Arena评测平台 (arena.ai). The article recommends using specific, detailed prompt templates focused on UI/screenshots, product labels, signage, interface/world knowledge, and widescreen formats to maximize success.

Leituras Relacionadas

Microsoft Identifies New Crypto Malware Targeting Wallet Addresses and Private Keys

In February 2026, Microsoft identified a new crypto clipper malware, dubbed Trojan/CryptoBandits.A, targeting Windows systems. The malware spreads via malicious shortcut files on USB drives and operates without a traditional installer or control servers by leveraging Windows Script Host and ActiveX to deploy a Tor proxy. Once active, it runs two modules: one for spreading and another for stealing information. The malware continuously monitors the clipboard for 12 or 24-word recovery phrases, Bitcoin/Ethereum private keys, and wallet addresses. When a user copies a wallet address, the malware silently swaps it with one controlled by attackers to divert funds. It also captures screenshots to gather information on wallet balances and user activity, sending data through Tor connections. Additional capabilities include remote code execution and persistence via scheduled tasks. Microsoft advises disabling auto-run features, restricting script interpreters and executable shortcuts from USB drives, and monitoring for suspicious activities like JavaScript execution, localhost:9050 proxy use, PowerShell screenshot capture, and clipboard monitoring.

TheNewsCryptoHá 21m

Microsoft Identifies New Crypto Malware Targeting Wallet Addresses and Private Keys

TheNewsCryptoHá 21m

No Sales Team, $20 Million in Revenue: How Did AI Employee Viktor Win Over 30,000 Companies?

The AI employee Viktor, developed by a team with DeepMind background, has achieved $20 million in annual revenue without a traditional sales team, serving over 30,000 companies. Its core innovation lies in positioning itself as a "Tier 3 AI Coworker" capable of "end-to-end execution and delivery of results," moving beyond the "draft and wait for human completion" model of typical AI assistants. Users can simply mention Viktor in Slack or Microsoft Teams using natural language commands, and it autonomously performs tasks like pulling sales data from a CRM, generating reports, or even cross-tool operations like creating board meeting PPTs by aggregating data from six different sources. Key to its growth is a pure Product-Led Growth (PLG) model, eliminating complex implementation cycles and per-seat licensing. Instead, it charges based on task credits or consumption, lowering the trial barrier with a $100 free credit offer and no credit card required. This enabled viral, bottom-up adoption within organizations. Viktor's interaction paradigm removes the barrier of prompt engineering, allowing non-technical employees to delegate complex workflows seamlessly. It also features proactive, automated task execution (e.g., overnight bookkeeping, scheduled reports) based on triggers, effectively embedding AI as an automated "process layer" within business operations. However, its expansion into Microsoft Teams—a platform with 320 million users—highlights challenges. Large enterprises require stringent IT compliance, security reviews (e.g., SOC 2), and governance, potentially hindering the frictionless, user-driven adoption that succeeded in Slack. Additionally, the "black box" nature of its autonomous decision-making raises concerns about operational risks, data integrity, and the need for robust audit logs and permission controls. Balancing efficiency gains with security and trust remains a critical hurdle for Viktor and similar AI agents aiming to become core enterprise infrastructure.

marsbitHá 59m

No Sales Team, $20 Million in Revenue: How Did AI Employee Viktor Win Over 30,000 Companies?

marsbitHá 59m

Interview with CoreWeave Co-founders: AI Demand Seems to 'Intensify' Every Day

An Interview with CoreWeave Executives: AI Demand Seems to 'Intensify' Every Day In an interview, CoreWeave executives highlight a structural shift in AI infrastructure demand. While GPU availability remains crucial, the primary bottlenecks are evolving to include powered data center shells, skilled labor (like electricians), and complex supply chain execution. They note that AI demand, particularly for agentic AI and reasoning models, continues to intensify daily, accelerating since Q1 2024. This demand is driving a need for more balanced infrastructure. CoreWeave is redesigning data centers to allocate more space for storage and CPUs alongside GPUs, with significant interest in Nvidia's upcoming Vera CPUs. The company, serving top AI labs and hyperscalers, emphasizes its client-driven model, building precisely to customer specifications. CoreWeave attributes its competitive edge to proven execution, performance, and a mature platform for AI deployment. Pricing is structured to pass component cost increases (e.g., for HBM memory) to customers, protecting margins. Looking ahead, they anticipate Vera Rubin platform deployments to begin meaningfully in late 2025, with a major ramp throughout 2027, mirroring the Blackwell (GB) series rollout pattern. The competition is shifting from merely acquiring chips to holistic engineering and delivery capability.

marsbitHá 1h

Interview with CoreWeave Co-founders: AI Demand Seems to 'Intensify' Every Day

marsbitHá 1h

Manus Buyback Plan Emerges: Chinese Investors Plan to Repurchase Equity with $2 Billion, Path to Hong Kong IPO Becomes Clearer

According to a report by The Information, early Chinese investors of Manus, including Tencent, Sequoia Capital China, and ZhenFund, are planning to repurchase the company from Meta for $2 billion—the same price Meta paid in its acquisition last December. This move is a direct response to the Chinese government's prohibition of the foreign acquisition in April. As part of the repurchase plan, Manus is considering establishing a Sino-foreign joint venture within China. This structure is seen as a way to ensure regulatory compliance for its Chinese investors and to pave the way for a future IPO in Hong Kong. Notably, U.S. investor Benchmark will not participate in the buyback, which will concentrate ownership even more among Chinese capital. Since its acquisition by Meta, Manus's business has grown rapidly, with its annualized revenue run rate reportedly increasing four-to-fivefold to $400-$500 million in roughly six months. This strong growth underpins the investors' willingness to repurchase at the original price. Financially, the forced unwinding of the deal may benefit the early investors, allowing them to regain equity at a cost far below the company's current implied valuation, with the added prospect of an independent future listing. However, specific terms of the repurchase, including funding proportions and the joint venture's equity structure, are still under negotiation. This "repurchase-joint venture-Hong Kong IPO" approach could serve as a reference model for other Chinese AI startups navigating cross-border M&A regulations.

marsbitHá 1h

Manus Buyback Plan Emerges: Chinese Investors Plan to Repurchase Equity with $2 Billion, Path to Hong Kong IPO Becomes Clearer

marsbitHá 1h

STRC Loses Peg by 11%, Can Strategy's Perpetual Motion Machine Keep Running?

The article discusses the significant and concerning depegging of MicroStrategy's (MSTR) preferred stock, STRC. Designed to trade near its $100 target par value, STRC has recently fallen sharply, reaching a low of $83.26 and closing at $88.59, representing an over 11% discount. STRC is a core component of MicroStrategy's financial strategy. As a perpetual preferred stock, it allows the company to raise capital through an "at-the-market" (ATM) issuance program without diluting common shareholders (MSTR). This capital is primarily used to purchase Bitcoin, creating a "capital flywheel": issuing STRC → raising cash → buying BTC → increasing net assets → supporting STRC's value. The flywheel's operation depends on STRC maintaining its $100 price. To enforce this, MicroStrategy employs a dynamic dividend mechanism, recently raising the rate to 11.5% and increasing payout frequency. However, this has failed to halt the depegging, indicating market concerns extend beyond yield. Analysts cite two main reasons. First, technical factors like forced liquidations from leveraged arbitrage trades may have exacerbated the sell-off. Second, and more fundamentally, is waning confidence in MicroStrategy's financial resilience. A JPMorgan report highlighted the company's limited cash relative to its ~$1.7 billion annual dividend obligation, raising liquidity concerns. While MicroStrategy counters that its massive Bitcoin holdings provide decades of coverage, this argument relies on the potential need to sell BTC—a departure from its long-standing "never sell" narrative. The company's recent sale of a small amount of Bitcoin for "testing," despite being framed as minor, has intensified these fears. The persistent depegging threatens to cripple MicroStrategy's primary funding channel. If STRC remains discounted, the company's ability to fund further Bitcoin purchases weakens. Should cash reserves dwindle while financing is constrained, the market may increasingly price in the risk of MicroStrategy becoming a forced seller of Bitcoin to meet obligations. This shift from a major marginal buyer to a potential seller could pose significant downside risk to the broader Bitcoin market.

链捕手Há 1h

STRC Loses Peg by 11%, Can Strategy's Perpetual Motion Machine Keep Running?

链捕手Há 1h

Trading

Spot
Futuros

Artigos em Destaque

Como comprar BANANA

Bem-vindo à HTX.com!Tornámos a compra de Banana Gun (BANANA) simples e conveniente.Segue o nosso guia passo a passo para iniciar a tua jornada no mundo das criptos.Passo 1: cria a tua conta HTXUtiliza o teu e-mail ou número de telefone para te inscreveres numa conta gratuita na HTX.Desfruta de um processo de inscrição sem complicações e desbloqueia todas as funcionalidades.Obter a minha contaPasso 2: vai para Comprar Cripto e escolhe o teu método de pagamentoCartão de crédito/débito: usa o teu visa ou mastercard para comprar Banana Gun (BANANA) instantaneamente.Saldo: usa os fundos da tua conta HTX para transacionar sem problemas.Terceiros: adicionamos métodos de pagamento populares, como Google Pay e Apple Pay, para aumentar a conveniência.P2P: transaciona diretamente com outros utilizadores na HTX.Mercado de balcão (OTC): oferecemos serviços personalizados e taxas de câmbio competitivas para os traders.Passo 3: armazena teu Banana Gun (BANANA)Depois de comprar o teu Banana Gun (BANANA), armazena-o na tua conta HTX.Alternativamente, podes enviá-lo para outro lugar através de transferência blockchain ou usá-lo para transacionar outras criptomoedas.Passo 4: transaciona Banana Gun (BANANA)Transaciona facilmente Banana Gun (BANANA) no mercado à vista da HTX.Acede simplesmente à tua conta, seleciona o teu par de trading, executa as tuas transações e monitoriza em tempo real.Oferecemos uma experiência de fácil utilização tanto para principiantes como para traders experientes.

141 Visualizações TotaisPublicado em {updateTime}Atualizado em 2026.06.02

Como comprar BANANA

Discussões

Bem-vindo à Comunidade HTX. Aqui, pode manter-se informado sobre os mais recentes desenvolvimentos da plataforma e obter acesso a análises profissionais de mercado. As opiniões dos utilizadores sobre o preço de BANANA (BANANA) são apresentadas abaixo.

活动图片

Categorias popularesright-arrow

Etiquetas Popularesright-arrow