Complexity Defaults High. Don't Let It Invent the SKU.

OpenAI’s image guide now documents GPT Image 2.6. The family has two Image API models: gpt-image-2.6-sunburst for detailed visual work and gpt-image-2.6-flare for fast iteration. Both take a complexity field: low, medium, or high. Omit it and both models use high. Complexity is how much the model plans before it renders. It is not a lock on the product. OpenAI’s own paper-cut football examples from the same prompt move the trophy, the uniforms, and the “CHAMPIONS” sign as complexity rises. Higher complexity can help dense text and layouts. It can also add latency and unwanted changes. Quality is a separate setting. Catalog and campaign stills still start from a photograph you already trust. Leave type empty. Overlay real words last. FreeGPTImg is a GPT Image 2 workspace: 20 daily free credits, no credit card required. Do not stamp 2.6, Flare, Sunburst, or complexity:high on a GPT Image 2 JPEG, and do not export the high-complexity cousin as the product.

Draft in FreeGPTImg. Stay on GPT Image 2 when that is the label on the screen. Recrop a feed frame in AI social media ads. Keep materials honest in AI product photography. Put one unit on white in the Amazon product photo generator. Change only the field in white background product photo. Change only one noun from a locked hero in image to image. Start a net-new concept in text to image only when you do not yet have a photo.

What the job actually is

OpenAI’s image guide still splits the work: generate from a prompt, or edit an existing image, including partial edits. GPT Image 2.6 sits on that Image API path. The model pages list /v1/images/generations and /v1/images/edits as supported. They list Responses as not supported. Do not send a 2.6 ID into a Responses action thread and expect the same handles as 2.5. The same guide still says the examples below the 2.6 section use GPT Image 2.5. Write the ID that actually ran.

Both 2.6 models support complexity control, markup-guided edits, SVG output, and animated GIF output through the Image API. Markup is still a circle on a transparent PNG, not a new bottle. SVG is for icons and illustrations with editable paths. GIF is a loop you describe in the prompt. None of those formats is Amazon MAIN. None of them is a Meta feed still you overlay later.

complexity is the unused handle. Low can suit a focused edit: change a background, change a sign. High can help dense text, complex layouts, and relationships between objects. Omit the field and you get high. There is no separate fee for changing it. There is a cost in pixels. OpenAI is explicit: higher complexity can increase latency and introduce unwanted changes, so compare results for your task. quality still controls rendering quality. Do not confuse the two. high complexity is not high quality, and auto quality is not a pinned grade on the pack.

Amazon’s product image help still treats MAIN as a realistic photo of the item you sell, on a pure white field (RGB 255, 255, 255), without extra graphics or type. A high-complexity cousin still is not MAIN. Meta’s image-ad guide for Facebook Feed currently recommends JPG or PNG, a 4:5 frame, and 1440×1800 as a working resolution, with a 600 px minimum width. That is a crop target, not permission to bake a “CHAMPIONS” sign into pixels you did not lock.

The cheap failure today is not “I did not read the 2.6 page.” It is “I omitted complexity, the model planned a denser scene, and now the grid disagrees with the box.”

Name the still before you generate

Weak sets fail because one 2.6 request tries to be a concept still, a high-complexity layout, MAIN, a 4:5 ad, an SVG mark, a nine-frame GIF, and the signed headline. Pick one job.

Job	What you export	Keep	Change one thing	Wrong move
Concept still	A mood board or placeholder	One idea	Nothing else	Feeding it back as the product
Locked catalog still	The SKU you already ship	Shape, color, label, logo, pack size	Nothing, or light only	Starting from a generated JPEG
White MAIN draft	Same SKU, square, opaque white	The upload	Field only	Chat chrome, extra unit, restyled colorway
4:5 feed still	Same SKU, empty lower or top band	Silhouette, palette	Crop and empty band	Asking high complexity to “make the scene richer”
9:16 story still	Same SKU, empty lower third	The object	Crop only	Status bars, prices, “2.6” badges
Focused edit	Same locked photo, complexity: "low"	Everything the prompt does not name	One noun: field or sign	Omitting complexity so it defaults high
Dense layout draft	A poster or infographic concept	The idea only	Nothing about a real SKU	Shipping it as the carton
SVG mark	A vector icon or wordmark, output_format: "svg"	Paths, not a photo	Nothing about MAIN	Editing the SVG as a raster product shot
GIF loop	An animation, generations only	Motion you described	Nothing that must match a listing	Using a 9-frame lantern as the pack shot
Offer overlay	Don’t generate it	Empty type	Editor, last	“SALE”, “$29”, “Shop now” in the prompt

A text-to-image pass invents. An image-to-image pass should borrow. If you already paid for a pack shot that matches the listing, that file is truth. A 2.6 high-complexity pass is extra planning. It is not a second product.

Checklist before anything leaves the thread

Use this sequence. Do not paste a campaign brief onto omitted complexity and expect the export to be the SKU you used to ship.

1. Name the job. Concept still, catalog still, MAIN draft, feed still, story still, focused edit, dense layout, SVG, GIF, or overlay. One file, one job.
2. If it is a catalog, ad, or listing still, start from a photograph. The locked JPEG of the item you sell. If you do not have that file, stop and photograph it. Do not invent a bottle because high complexity packed the scene.
3. Write keep / change / do-not in five lines. Keep silhouette, color, label, logo, and pack size. Change only one noun: crop, field, empty band, or one sign. Do not add a second SKU, a restyled colorway, or a percent.
4. Name the engine on the screen. If FreeGPTImg shows GPT Image 2, write gpt-image-2. If an API log says gpt-image-2.6-sunburst or gpt-image-2.6-flare, write that ID. If you set complexity, write low, medium, or high into the filename too. Do not relabel either file 2.6-approved.
5. Lock the still outside 2.6 first. Run the keep / change / empty-type loop on GPT Image 2. Only then send that PNG into a 2.6 edit, and only with complexity: "low" if the file must stay the SKU.
6. Do not omit complexity on a shipping still. Omit means high. High plans more. High can move the sign, the uniform, and the trophy in OpenAI’s own examples. If the photograph is already truth, pin low and name one noun. If you do not yet have a photo, pin the complexity you meant and keep that file in the concept folder.
7. Ask for empty type. Prices, percents, star ratings, “Shop now,” legal lines, and model names never belong in the generator if they must be exact. Overlay them in an editor after a person signs the words.
8. Leave production room. Product in the middle third. Quiet field. Empty band where the headline will sit. Do not fill every edge with planned extras, arrows, or a “CHAMPIONS” banner you did not lock.
9. Generate one still, then multiply. One locked pass first. File names like kettle-4x5-empty-band-gpt-image-2.png. Never kettle-final-complexity-high.png for a listing. An SVG is a mark. A GIF is a loop. n greater than 1 is a concept set, not a listing gallery.
10. Inspect the subject before the overlay. A rewritten mark, a floating “high”, a ribbon that says “SALE”, or a tag that says “4.9” are stop conditions. Hold the still next to the real pack. If a stranger would think this brand shipped a 2.6 edition, discard it.
11. Only then overlay. In an editor, set the real headline and only the claims a manager signed. Recrop to the live placement size after the pixels match. Do not ask high complexity to spell the price. Do not ask Sunburst to invent a lid that “reads cleaner in a denser scene.”

Prompt formula for the locked still

GPT Image 2 catalog or ad still, not a 2.6 high-complexity generate and not MAIN unless named.
Start from the uploaded product photo.
Keep: exact silhouette, color, label, logo, pack size.
Change only: [crop / field / empty headline band / one noun].
Empty type. No prices, percents, ratings, model names, watermarks, or chat chrome.
Do not add: extra SKUs, restyled colorways, “Shop now”, celebrity faces, children.

Prompt examples

Locked catalog still. “Image-to-image: keep this exact matte-black kettle, spout curve, handle, and printed mark. Even studio light. Empty type. Not a 2.6 reshoot, not MAIN unless named.”

White MAIN, not a high file. “Image-to-image: keep this exact SKU. Change only the field to opaque pure white. No badge, no extra unit, no planned extras.”

4:5 feed, empty lower band. “Image-to-image: keep this exact kettle. 4:5 crop, product in the upper two thirds, empty lower band for later overlay. Quiet kitchen, even light. No prices, no 2.6 chrome, no second kettle.”

9:16 story, empty lower third. “Image-to-image: keep this exact mug glaze and printed mark. 9:16, empty lower third. No status bar, no ‘Shop now’, no model watermark.”

Focused edit after the still is locked. “Use this product photograph exactly as uploaded. complexity: low. Change only the lid finish to the same matte black as the body. Do not restyle, recolor, or redraw the rest of the SKU. Empty type.”

Do not generate this. “complexity high, make the scene richer, new copper glaze, ten colorways, CHAMPIONS banner, Instagram CTA Shop now $29, 4.9 stars, guaranteed Amazon approved, cousin lid that looks cleaner at max, SVG of the bottle, 9-frame GIF of the pour.”

What to tell an explicit 2.6 edit after the still is locked

If you already have the GPT Image 2 PNG, a 2.6 Image API edit on gpt-image-2.6-sunburst can still be useful. Keep the instruction short and negative about the product. Pin complexity: "low". Pin quality. Put the photograph in the edit first. Recheck the live model page: 2.6 is generations and edits, not Responses.

Use this product photograph exactly as uploaded. Do not restyle, recolor, relight, or redraw the SKU.
Keep labels, logo, silhouette, and pack size accurate.
Change only: [one noun].
Leave all type empty. Do not invent prices, percents, or CTAs.
complexity: low
quality: high

That is a focused edit. It is not a second generate of the object, and it is not a license to leave complexity on the default.

Review the pixels before they leave the overlay

Zoom to 400%. A planned still hides a rewritten label. An overlay will not hide it either.

• Subject honesty. Same silhouette, color, label, and pack size. If a stranger would think this brand shipped a 2.6 edition, discard the still.
• Job honesty. A concept still stays in the concept folder. A high-complexity export is a high-complexity export until you prove otherwise. MAIN stays white, one unit, no graphics. An SVG stays a mark. A GIF stays a loop.
• Engine honesty. Write the model you used into the file name. A GPT Image 2 still is GPT Image 2. A 2.6 Sunburst edit still is Sunburst plus complexity:low. Do not stamp 2.6 on the FreeGPTImg JPEG.
• Empty type. Prices, percents, ratings, store chrome, and model names stay empty of anything you need people to read. Planning is not a headline.
• Handle honesty. low, medium, or high. Not omitted on a shipping file. Quality is not complexity. SVG output does not go back into image edits. GIF output is generations only. If you needed a new object, you did not need a focused edit.
• People and file. No recognizable generated faces you cannot clear. No children. PNG or JPEG, RGB, quiet field, no QR. Preview at feed size and at the size the catalog actually renders.
• Provenance. A 200 from the Image API is not a license, not a marketplace approval, and not a signed offer. Recheck Amazon’s product image help and Meta’s ads guide before you tell a teammate the file is ready.

Limitations you should not skip

A high-complexity export is not a signed discount, not a commercial-rights grant, and not MAIN. FreeGPTImg does not run GPT Image 2.6, complexity, Flare, or Sunburst. The default loop here is GPT Image 2 stills. Do not generate medical seals, “#1 2026,” store logos, or sale badges unless those claims are already on the real label and a manager signed them.

OpenAI’s own limitations still apply: complex prompts can take up to two minutes; text in the image can still miss placement and clarity; recurring brand marks can drift across generations; structured layouts are harder than they look. Treat those as reasons to lock the photograph first, not as a dare to raise complexity.

The 2.6 model pages currently support image generation and image edit. They do not list Responses. SVG needs output_format: "svg" on generations, n of 1, and is not a supported raster input for edits. GIF needs output_format: "gif" on generations; frame_count is 2–64 (default 9) and frame_duration_ms is 20–2000 (default 200). Neither format is a listing photo. Recheck the live docs; model IDs, complexity defaults, and output formats move.

Models invent letters, digits, and product cousins when you mention Instagram, a dollar figure, “richer scene,” “more premium,” or “make ten.” Shorten the brief. One SKU. One noun. Empty type. Default high is extra planning, not a backup of the PNG you already had. If the locked raster is honest, recrop that. If it is not honest, rephotograph. Do not ask a denser plan to remember a pack you never locked.

FAQ

What does complexity do on GPT Image 2.6?

It controls how much the image model plans before rendering. Supported values are low, medium, and high. If you omit it, both Sunburst and Flare use high. Low can suit a focused edit. High can help dense text and layouts, and can also add latency and unwanted changes. Quality is a separate setting. Recheck the live image-generation guide.

Should I leave complexity omitted for catalog stills?

No. Omit means high. High is a plan, not a lock. If the photograph is the SKU you sell, set complexity: "low" and tell the model not to restyle the pack. If you do not yet have a photo, set the complexity you meant and keep that file in the concept folder.

Does FreeGPTImg run GPT Image 2.6, complexity, Flare, or Sunburst?

No. FreeGPTImg is a GPT Image 2 workspace: 20 daily free credits, no credit card required. Use it to lock the still. Write the name on the screen into the filename.

Can I use a high-complexity still as Amazon MAIN or a Meta ad?

Not if it changed the product, and not because the API returned 200. MAIN is a realistic photo of the item you sell, usually on pure white, without extra graphics or type. A restyled glaze, a denser scene, or baked “Shop now” type fails that job. Lock the SKU first. Empty type. Overlay last. Recheck Amazon’s product image help and Meta’s ads guide.

Why did the CHAMPIONS sign move when I raised complexity?

Because OpenAI’s own 2.6 examples show the same prompt producing different trophy, uniform, and sign placement as complexity goes from low to high. Planning is not a promise the pack stays still. Pin low when the photograph is already truth.

Should I lock stills in 2.6 or outside it?

Outside it, if the file might ship as catalog, MAIN, or an ad. Photograph the item. Run the keep / change / empty-type loop on GPT Image 2. Then, if you still need a 2.6 edit, send that PNG with complexity: "low" and tell the tool not to touch the rest of the SKU. Do not send 2.6 into Responses; the current model pages do not list that endpoint.

Sources

• OpenAI: Image generation
• OpenAI: GPT Image 2.6 Sunburst
• OpenAI: GPT Image 2.6 Flare
• Amazon: Product image requirements
• Meta: Image ads guide

Lock the still before you leave complexity on high

Pick the PNG that already matches the item you sell, not the JPEG a 2.6 pass invented because complexity defaulted to high. Write five lines: keep, one noun, empty type, overlay later. Open FreeGPTImg, stay on GPT Image 2 when that is the label on the screen, recrop a feed frame in AI social media ads, keep materials honest in AI product photography, put one unit on white in the Amazon product photo generator, change only the field in white background product photo, run the edit in image to image, start a net-new concept in text to image only when there is no photo, and only then overlay the signed words — or send the locked PNG into a 2.6 Image API edit with complexity: "low", never omitted high on a shipping file, and tell the tool not to touch the SKU.