OpenAI shipped gpt-image-2 on April 21, 2026, and it is the first image model that actually belongs in a production pipeline. Not because the pictures are prettier. Because the API finally does the things I kept wanting the old one to do.

What actually changed

Three things, and you can ignore the rest of the announcement.

It reads and writes legible text. OpenAI claims ~99% accuracy on typography, including CJK and right-to-left scripts. That is a big deal if you have ever tried to generate a product label or a slide deck header and gotten cursed runes back. The old model was a pixel painter. The new one is a pixel painter that can spell.

It thinks before it draws. There is a reasoning pass baked into the model now, a “think about the scene, then render” step. You do not have to configure it. You do not pay a thinking-mode surcharge on the standard API call. It just converges faster. Prompts I used to iterate on three or four times now land on the first or second try.

It edits images. Real editing, not “here’s a new image that vaguely resembles your old one.” You pass in a picture and a description of what you want changed, and the rest stays put. This is the capability that makes it worth wiring into a pipeline.

The minimum viable call

If you have Node 20+ and an OpenAI API key, this is the whole thing:

curl https://api.openai.com/v1/images/generations \
  -H "Authorization: Bearer $OPENAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-2",
    "prompt": "A cinematic 4K landscape of an AI data center at twilight",
    "size": "1536x1024",
    "quality": "high"
  }'

You get back base64 image data in data[0].b64_json. Write it to a file. Done.

One thing that tripped me up: do not send response_format. The docs say it is not supported, and they mean it. The API rejects the request with HTTP 400. All responses come back base64 only. If you want URLs, host them yourself.

The edit endpoint is the real unlock

Here is the image I generated first, a photorealistic 4K data center at twilight:

Now here is the same image after a single edit call: “replace the twilight clouds above the servers with a dramatic aurora borealis, ribbons of emerald green, magenta, and electric teal. Keep everything else unchanged.”

No mask. No Photoshop. The server racks are in the same positions. The orange horizon is preserved. The blue light trails between the racks still flow toward the vanishing point. Only the sky changed.

Try doing that with gpt-image-1. You cannot. The /v1/images/edits endpoint existed before, but the results it gave you were not the kind of thing you shipped to production. This is the feature I was waiting for.

The endpoint accepts multiple reference images, which you address inside the prompt as “image 1” and “image 2” for compositing. Style transfer, product placement, character relocation, all one API call.

Three creative moves the marketing post does not tell you about

Aspect ratios no one else gives you. The preset list is short, but size accepts custom values. Both dimensions must be multiples of 16, max edge 3840, aspect ratio up to 3:1, and total pixels between 655,360 and 8,294,400. That range covers Twitter cards at 1200×628, Instagram stories at 1080×1920, blog heroes at 1920×1080, and full 4K landscapes at 3840×2160. No cropping, no upscaling, no extra tooling.

One caveat the docs hide: anything above 2,560×1,440 is officially “experimental.” It works. I generated a 4K image for this post. But OpenAI is not promising SLA on it yet, so budget for occasional failures in production.

Batch consistency. The n parameter goes up to 8, and the model keeps characters and objects consistent across the set. For a product shot or a children’s book page, one call gives you eight variations that actually share visual DNA. Eight variations at medium quality costs about thirty cents. That is a lot cheaper than eight separate prompt-engineering sessions.

Reasoning as a debugging tool. Because the model thinks before drawing, iteration feels different. Vague prompts still produce vague images, but specific prompts land harder. I stopped writing six-paragraph mega-prompts and started writing three-sentence scene descriptions with hex colors and composition direction. The output got better.

Gotchas

Things I wish the docs had told me louder:

• No transparent backgrounds. If you need a PNG with alpha for icon work, you still reach for gpt-image-1. Route by use case.
• Masks are prompt-guided, not pixel-exact. If you are coming from Stable Diffusion, this will feel wrong. The mask tells the model which region to focus on. The model decides how to blend.
• C2PA watermarks are on by default. Every image ships with provenance metadata. Useful for trust, relevant if you were hoping to redistribute without attribution.
• Streaming partials cost extra. Each partial_images frame adds 100 output tokens. Fine for prototyping a UI. Expensive at scale.
• Pricing is per token. $8 per million input tokens, $30 per million output tokens, with the usual caching discount. A medium 1024×1024 lands around four cents. A high-quality 4K lands near eighty cents. The calculator on the docs page will save you some math.

Where to take it

If you are already using image generation in an app, switching is low-risk. Change the model string, drop response_format if you were sending it, and audit for transparent-background assumptions. Your latency will drop. Your text will be legible.

If you are not using image generation in an app, the editing endpoint is the reason to start. Every product that has ever wanted “make this photo match our brand” can now do it with three lines of code.

I rebuilt my cartoon pipeline around the new editing flow in about an afternoon. The first draft of this post was going to be a benchmark comparison. Then I looked at the aurora edit and realized there was nothing to benchmark. Either your tool can do it or it cannot.

gpt-image-2 can.

Claude Code might be the most significant event of 2025 in the field of AI. For me, it changed the course of the year and led me to unthinkable places. I followed every twist and turn and stayed on top of the wave. What an amazing tool that delivers on its promise. It amplifies and scales your ability, keeps you moving forward, and leads to some serious compounding.

An overlooked impact of AI is its compounding nature. From the first prompt, where everything is wrong. But, slowly, you learn, you figure out how to harness. At some point, you forget about the first hallucination, and you are in a new spot. When I talk to others, I can immediately tell where they are on the curve by how they talk about these tools. A scary thing, though, is how far behind you can get, and that you might not be able to catch up to those who have started. If you do anything this month, go down the bumpy road and get on the other side.

As a baseline, here’s the AI-generated image from two years ago, created with DALL-E 3.

The original DALL-E 3 photo looks kind of crazy when you look back. I would never garnish a drink with a grapefruit wedge; it would have been a peel.

To remind you of the cocktail recipe for The Merry Manhattan:

Print

The Merry Manhattan

Recipe by Hans Scharler

Prep time

5

minutes

The Merry Manhattan is a festive twist on the classic Manhattan cocktail, perfect for holiday celebrations. This elegant drink features a rich amber hue, achieved by blending rye whiskey with sweet vermouth. The traditional flavor is enhanced with a unique addition of smoked cherries, adding a subtle, smoky sweetness. A sprig of rosemary infuses the cocktail with a fragrant, herbaceous aroma, invoking the essence of winter. The drink is served in a rocks glass containing a large ice chunk. The finishing touch is a gracefully twisted grapefruit peel, adding a citrusy zing and completing the cocktail’s holiday charm.

Ingredients

• 1 oz Carpano Antica Formula Vermouth

• 2 oz Whistle Pig Rye Whiskey

• 2 dashes Sour Cherry Bitters

• 2 dashes Peychaud’s Bitters

• Fresh cherries (for smoking)

• Fresh rosemary (for smoking)

• Grapefruit peel (for garnish)

• Ice

Directions

• Prepare the Smoke:
  Gather a few fresh cherries and a sprig of rosemary.
  Using a kitchen torch, gently torch the rosemary and cherries until they start to smoke. Be careful not to burn them.
  Immediately cover the smoking rosemary and cherries with a rocks glass to trap the smoke inside. Let it sit for a minute to infuse the glass with the smoky aroma.
• In a mixing glass, combine 1 oz of Vermouth and 2 oz of Rye Whiskey.
• Add two dashes each of Sour Cherry Bitters and Peychaud’s Bitters.
• Fill the mixing glass with ice and stir well to chill and dilute the cocktail.
• Add an ice chunk to the smoked rocks glass.
• Strain the stirred cocktail into the smoked rocks glass.
• Finish and Garnish:
  Take a grapefruit peel and express (squeeze) its oils over the drink.
  Use the grapefruit peel as a garnish.
  Use one of the smoked cherries as a garnish.
  Use the sprig of rosemary as a garnish.

Notes

• Enjoy the smoked cherries as a treat, or save them for a future cocktail garnish.

Here’s the first try with Nano Banna Pro. I gave it the recipe card along with the prompt to create a realistic photo for a Christmas or New Year’s party setting.

I am still not in love with the photos. They are better for sure. Let me try a new approach. I am going to just send a link to the blog post and ask for the photos again.

This time it was much better. I like that it used a fancy cocktail cherry, like a Luxardo cherry, instead of one with a stem.

I found this an interesting way to visualize how a drink can come together. I like inventing my own cocktails for parties. This gives me a way to experiment with the visual presentation.

Now, Nano Banana Pro has way more capability than DALL-E 3 had, so I can do more things. I can make process diagrams for the recipe card. Let’s try that.

Now, we are talking.

A few takeaways:

• Models are improving rapidly.
• You can do something new with Generative AI models.
• And, the power of AND. I started with a photo, then pivoted to a recipe card diagram. I could keep anding. I could make it a YouTube video script. I could make it a series of cocktails. This is the most critical takeaway for 2026. It is not just about doing one thing more efficiently; it is about doing more things than you could before.

Buckle up. You might need a drink in 2026.

Gemini 3 Pro

Google went a different direction. Gemini 3 Pro is all about reasoning, multimodal inputs, and that million-token context window.

The benchmark numbers are wild. It hit 91.9% on GPQA Diamond. On ARC-AGI-2, the abstract reasoning benchmark, it scored 31.1% (and up to 45% in Deep Think mode). That is a huge leap over previous models. On LMArena it took the top ELO spot.

If your work is heavy on reasoning, vision, video, or you need to throw massive context at a problem, Gemini 3 Pro is built for that.

Claude Opus 4.5

Anthropic announced Opus 4.5 on November 24, 2025. They are calling it the best model in the world for coding, agents, and computer use. Bold claim.

On their internal engineering benchmarks, Opus 4.5 scored higher than any human candidate ever on their take-home exam. It also delivers higher pass rates on tests while using up to 65% fewer tokens than Sonnet 4.5. That efficiency piece matters if you are building agents that run for hours.

The pitch is clear: if you care about code, automation, and not burning through tokens, Opus 4.5 is the one.

How They Compare

Opus 4.5 is aimed at engineers building agents and writing code. It is optimized for efficiency. Gemini 3 Pro is aimed at everything else: reasoning, multimodal, long context, general purpose.

Both are frontier models. Both are available on major clouds and APIs. The honest answer is you might end up using both depending on the task.

The Real Point

The meta-point is not which model wins. The point is that two frontier models landed in the same week, both pushing hard on different axes. Reasoning, coding, agents, vision, efficiency. The pace of improvement is nuts.

If you are building with AI right now, the table is full. Pick your model, match it to your task, and start experimenting. There has never been a better time.

What excites me about Claude Opus 4.5?

While I was writing this post, a friend texted me to ask what excited me about Claude Opus 4.5. I spent a couple of hours with it today, and I would have to say… tool search.

Tool search lets Claude discover and load tools on-demand instead of pre-loading every definition at the start.

Here is how it works. You provide a catalog of tools to the API with names, descriptions, and input schemas. You mark most of them with defer_loading: true, which means they stay out of the model’s context until needed. Then you include a tool-search tool in the list. When Claude needs a new capability, it searches, finds the right tool, and only then does that tool get loaded into context.

There are two pain points this solves… expensive tokens and picking the right tool.

When you load dozens of tools upfront, the definitions alone eat up thousands of tokens. That is space you could be using for reasoning, tool outputs, and user messages. With tool search, you load only the search tool and maybe three to five relevant definitions. The overhead drops significantly.

With large libraries of tools, the model can struggle to pick the right one, especially when names or parameters are similar. The search step narrows the candidates to tools that actually match the task.

The latest agent IDE is called Antigravity by Google. It is based on their top coding model, Gemini 3, and their own agent framework. It is brand new, and people are crushing it with requests; they are making frequent updates and fixing bugs.

I noticed when I started a new chat for my project, Antigravity thought about my request and opened a “brain” file. It looks to be some markdown that Antigravity is managing as I work on the project. This seems like a smart idea and might be where Google can differentiate itself from the competition.

As they say, AI Engineer Code is by invitation only. They have hand-selected the top builders and leaders in AI Engineering for a highly curated experience. The team behind the event also helped organize meetups and hackathons happening around the city.

Some interesting sessions (from descriptions) that I am looking forward to:

• 2026: The Year the IDE Died by Steve Yegge, Sourcegraph, and Gene Kim, IT Revolution
• Autonomy Is All You Need by Michele Catasta, Replit
• Don’t Build Agents, Build Skills Instead by Barry Zhang and Mahesh Murag, Anthropic
• Research to Reality with Google DeepMind by Kat Kampf and Ammaar Reshi, Google

Some of the AI Engineer Code events will be live-streamed on YouTube. Check out the video here:

I will share my thoughts on the experience and key takeaways after the events, seminars, and workshops. Should be awesome, profound, scary, and awesome.

Side note: On the AI Engineer Code site, there was a cool wave ASCII art animation that showed the AIE logo, NYC landmarks in an animated wave. I used Claude to make me one.

Riding the wave is a wild ride.

• gpt-5 will be designed for logic and multi-step tasks.
• gpt-5-mini is a lightweight version for cost-sensitive applications.
• gpt-5-nano is optimized for speed and ideal for applications requiring low latency.
• gpt-5-chat is designed for advanced, natural, multimodal, and context-aware conversations for enterprise applications.

As large language models and other GenAI models advance, you need to shift your mindset on how to use them. Here’s how I leverage advanced models and how I have shifted my prompting.

1. Structural Integrity: Crafting Clear and Organized Prompts

Think of your prompt as a blueprint for the AI model. A well-structured prompt ensures the model understands your intent and constraints effectively.

• Guardrails and Edge Cases: Don’t just describe the ideal scenario; also consider the exceptions. Anticipate potential issues or deviations and explicitly instruct the AI on how to handle them. For example, if you’re asking for a summary of articles, specify what to do if an article is paywalled or inaccessible. This proactive approach leads to more robust and reliable outputs.
• Strategic Context Positioning: Where you place information within your prompt can influence the AI model’s attention.
  • Front-load critical instructions (first 10%): Make your core request and essential rules immediately clear.
  • Middle ground for context and data: Provide necessary background information, examples, or data in the central part of the prompt.
  • Reinforce key constraints at the end: Briefly reiterate any crucial limitations or desired formats to leave a lasting impression.
• The Power of “Don’t”: Surprisingly, providing negative examples – explicitly stating what you don’t want the AI model to do can be more effective than solely focusing on positive examples. By illustrating failure modes, you guide the model away from undesirable outputs.

2. Evidence-Based Techniques: Leveraging the Model’s Strengths

Move beyond simple requests and employ techniques that tap into the AI model’s reasoning and self-awareness:

• Test for Self-Consistency: For critical outputs, ask the AI model to generate multiple responses to the same prompt. Analyzing the consistency across these responses can give you a better gauge of the reliability of the information.
• Unleash “Program of Thought”: For tasks involving logic, math, or code generation, explicitly instruct the model to “solve this by writing a program” or “show your work step-by-step using calculations.” This encourages the AI to leverage its tool-use capabilities for more accurate results.
• “Plan and Solve” for Complex Tasks: Before asking the AI model to execute a complex task, request it to first outline a step-by-step plan. This allows you to review the proposed approach, identify potential flaws in its logic, and guide it towards a more effective strategy before the final output is generated.

3. The Art of Metaprompting: Talking to the AI Model About Itself

A new concept that you should consider is metaprompting. This is you prompting the AI model to reflect on its capabilities and limitations. Since advanced models possess a significant understanding of their workings, you can leverage this knowledge to improve your results.

• The Self-Improvement Loop: Simply ask: “Here’s my current prompt: [your prompt]. How would you improve this prompt to get better results from you?” The AI model can often provide valuable suggestions for clarity, specificity, or even the inclusion of techniques you haven’t considered.
• Checking for Uncertainty: Proactively ask: “What parts of this request are unclear or ambiguous? What assumptions are you making? What additional information would help you execute this prompt with more accuracy?” This can help uncover potential misunderstandings and prevent overconfident, yet inaccurate, responses.
• Discovering Hidden Potential: Inquire: “How would you approach this if you had no constraints? What would be your ideal process? What tools or information would help you?” This can reveal the model’s full potential and suggest innovative approaches you might not have thought of.
• Demanding Explainability: Ask: “Explain your reasoning step by step. What parts are you most or least confident about?” Understanding the AI’s thought process can help you diagnose issues and build trust in its output.
• The Socratic Approach: Use probing questions like “Why did you choose that approach?” or “What alternatives did you consider?” to encourage deeper reflection and uncover underlying assumptions in the AI’s reasoning.

I typed these words into a text box: Elephants Wear Fancy Hats to a Party

This is the video that was produced two minutes later.

Try the latest AI models from Google at https://gemini.google.com/.

What are you experimenting with?