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.

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.

I will share my experience generating art with AI. I am not an artist, but I have ideas and access to a web browser that has access to power Generative AI tech. There are many platforms to experience, but I recommend that you pick one, try out prompts, iterate, and share. I picked Stable Diffusion.

Stable Diffusion is an advanced generative model for creating images, inspired by the diffusion process in physics. When given a text prompt, the model generates an image based on the content and themes described in the prompt.

M. C. Loves Robots

I love the artstyle of M. C. Escher. With Generative AI, I can now get an idea of what Escher would create if he was obsessed with robots.

Let’s unpack the prompt used to generate this image

The prompt “M. C. Escher lithograph, robots, gears, lights, wires” consists of several components that give the model an idea of what elements to include in the generated image.

1. “M. C. Escher lithograph” – This part of the prompt instructs the model to generate an image inspired by the artistic style of M.C. Escher, a Dutch graphic artist known for his mathematically inspired works that feature impossible objects, optical illusions, and intricate patterns.
2. “robots” – This keyword tells the model to include robots as one of the central subjects in the image. The robots may be designed in a way that reflects Escher’s style or is inspired by his works.
3. “gears” – The model should incorporate gears into the image, possibly as part of the robots’ mechanisms or as a separate visual element.
4. “lights” – This term suggests that the image should have a focus on lighting, either by using the lights as a part of the composition or emphasizing the way they interact with the other elements in the scene.
5. “wires” – The presence of wires indicates that the generated image should have a somewhat intricate or complex appearance, with wires possibly connecting various elements or being part of the robots.

When the Stable Diffusion model receives this prompt, it will generate an image that combines these elements. The resulting artwork might be an intricate scene of robots, gears, and wires, all rendered in the distinctive style of M.C. Escher, with a focus on the interplay of light and shadow. The specific details of the generated image will depend on the model’s interpretation of the prompt and its internal knowledge of the various elements mentioned.

Each word in a Generative AI prompt matters. You can get wild variations by changing out the words and experimenting.

Stable Diffusion Prompt Engineering Tips

Stable Diffusion is a groundbreaking generative AI art model that enables the creation of stunning and unique artwork. By providing the model with a carefully designed prompt, you can guide its creative process to generate art that aligns with your vision.

Be Specific and Clear

When writing prompts for Stable Diffusion, it’s crucial to be specific and clear in your description. This ensures that the model can accurately interpret your intentions and generate artwork that matches your vision. For example, instead of writing “a landscape,” provide more detail such as “a serene sunset landscape with a calm lake and a silhouette of a tree.”

Incorporate Emotions and Atmosphere

To create truly captivating art, consider incorporating emotions and atmosphere into your prompts. This can help the model generate art that evokes a certain mood or feeling. For instance, instead of writing “a cityscape,” try “a bustling cityscape with a lively atmosphere, filled with people enjoying their day.”

Experiment with Different Styles

Stable Diffusion is capable of producing artwork in various styles, from abstract to photorealistic. When crafting your prompt, consider specifying a particular artistic style or movement to inspire the model. For example, you could write “a vibrant, impressionistic painting of a Parisian street café.”

Combine Unusual Elements

One of the most exciting aspects of generative AI art is its ability to combine seemingly unrelated elements to create unique and thought-provoking pieces. To harness this potential, try writing prompts that incorporate contrasting or unexpected components. For example, “a surreal scene of a whale floating above a city skyline with hot air balloons in the background.”

Iterative Refinement

Don’t be afraid to iterate and refine your prompts. If the generated artwork isn’t quite what you envisioned, try adjusting the wording or adding more details to your prompt. It’s essential to experiment and adapt your prompts to hone your skills in guiding the model’s creativity.

Stable Diffusion offers an exciting avenue for artistic expression, and writing effective prompts is key to unlocking its full potential. By being specific, incorporating emotions, experimenting with styles, combining unusual elements, and refining your prompts, you can guide the model to generate visually stunning and emotionally evocative artwork. So go ahead, unleash your creativity, and let Stable Diffusion bring your artistic visions to life!

Use ChatGPT if you get stuck

A neat thing that you can do is use ChatGPT if you get stuck. Use AI to write AI prompts. Ask ChatGPT to give you ideas for prompts for an artist that you like. I always loved President Obama’s presidential painting by Kehinde Wiley. A robot is likely to be our future president and Kehinde will be tapped once again to paint the portrait. I used ChatGPT to help me generate ideas for this painting.

Enjoy the journey

We are going through some weird times. Enjoy the journey. Try out prompts and share the results on our Discord server. I would love to see what you come up with.