GPT Image 2.5 Sunburst is OpenAI’s precision-oriented model for image generation and editing. It accepts text and image inputs and produces image outputs. Use it when a creative task depends on carefully modifying existing material, working with visual references or refining a composition through several steps. The brief can create a new scene, transform an existing image or guide an edit to a selected region. Sunburst and Flare are separate editions: Sunburst emphasises editing precision, while Flare is positioned for fast everyday generation.
For a new image, describe the subject, setting, composition, lighting and style, then state the intended shape and use. For a product concept, request a glass lamp on a stone surface with soft side lighting, a clearly visible silhouette and open space for a headline. For an edit, name the change and the elements that should remain recognisable: replace the background, adjust the light or alter a material while retaining the selected composition. The direct producer generation and edit workflows permit a prompt up to 32,000 characters and one to ten results.
Reference editing can use up to sixteen input images. Give each image a role so the instruction distinguishes the subject to retain, an environment to use and a style to apply. The model can also work with a mask that indicates an area to change. The source and mask need the same format and dimensions, and the mask must carry an alpha channel; the guide gives a size requirement below 50 MB. With several source images, the mask applies to the first. It guides the requested change rather than guaranteeing a hard boundary, so include preservation instructions in the brief and inspect the surrounding area in the result.
Quality settings are low, medium, high, xhigh, max and auto, with auto as the default. Low can support a quick visual draft, while higher settings are useful to compare when fine detail matters. Quality is separate from output dimensions and file compression. Standard sizes include 1024 × 1024, 1536 × 1024 and 1024 × 1536 pixels, and automatic or custom dimensions are also available. Choose the shape around the final layout before deciding how much detail the image needs.
Custom dimensions must be multiples of sixteen pixels, neither edge may exceed 3840 pixels, the aspect ratio must stay between 1:3 and 3:1, and total area must lie between 655,360 and 8,294,400 pixels. These conditions apply together. A 3840 × 2160 landscape fits the upper area limit, while a long edge by itself does not make an arbitrary pair of dimensions valid. Resolutions above 2560 × 1440 are described as experimental. Keep important objects, text and useful negative space within the intended crop instead of relying on more pixels to resolve the composition.
PNG is the default output format; JPEG and WebP are also supported. Compression from 0 to 100 can be selected for JPEG and WebP. Transparent, opaque and automatic backgrounds serve different layout needs. For transparency, choose PNG or WebP so the result can retain an alpha channel. A transparent asset can then be placed in another composition, while an opaque image supplies its own background. Keep the actual completed file together with the brief and selected output settings.
OpenAI documents direct generation and editing as well as a conversational image-generation tool that can retain prior results and apply follow-up instructions. In that conversational workflow, the action can be automatic, forced generation or forced editing when an image is in context. Use the chosen image as the starting point for the next change rather than silently switching to another variation. The documented streaming workflow can deliver up to three partial images before the final result; a partial preview helps show progress but is not the finished asset. The manufacturer interface also supports batch processing.
Check the edited subject, lettering, object placement and the boundary of a masked change at the intended display size. GPT Image can still struggle with precise text positioning, structured layouts or consistency across recurring subjects and brand elements. A focused next instruction should identify the visible issue and the elements to preserve. Sunburst is useful for a precise, reference-led creative workflow whose final artifact can be assessed against a concrete visual brief, with higher quality, dimensions and transparency chosen for that artifact.
", "aboutText": "GPT Image 2.5 Sunburst is OpenAI’s precision-oriented model for image generation and editing. It accepts text and image inputs and produces image outputs. Use it when a creative task depends on carefully modifying existing material, working with visual references or refining a composition through several steps. The brief can create a new scene, transform an existing image or guide an edit to a selected region. Sunburst and Flare are separate editions: Sunburst emphasises editing precision, while Flare is positioned for fast everyday generation.\n\nFor a new image, describe the subject, setting, composition, lighting and style, then state the intended shape and use. For a product concept, request a glass lamp on a stone surface with soft side lighting, a clearly visible silhouette and open space for a headline. For an edit, name the change and the elements that should remain recognisable: replace the background, adjust the light or alter a material while retaining the selected composition. The direct producer generation and edit workflows permit a prompt up to 32,000 characters and one to ten results.\n\nReference editing can use up to sixteen input images. Give each image a role so the instruction distinguishes the subject to retain, an environment to use and a style to apply. The model can also work with a mask that indicates an area to change. The source and mask need the same format and dimensions, and the mask must carry an alpha channel; the guide gives a size requirement below 50 MB. With several source images, the mask applies to the first. It guides the requested change rather than guaranteeing a hard boundary, so include preservation instructions in the brief and inspect the surrounding area in the result.\n\nQuality settings are low, medium, high, xhigh, max and auto, with auto as the default. Low can support a quick visual draft, while higher settings are useful to compare when fine detail matters. Quality is separate from output dimensions and file compression. Standard sizes include 1024 × 1024, 1536 × 1024 and 1024 × 1536 pixels, and automatic or custom dimensions are also available. Choose the shape around the final layout before deciding how much detail the image needs.\n\nCustom dimensions must be multiples of sixteen pixels, neither edge may exceed 3840 pixels, the aspect ratio must stay between 1:3 and 3:1, and total area must lie between 655,360 and 8,294,400 pixels. These conditions apply together. A 3840 × 2160 landscape fits the upper area limit, while a long edge by itself does not make an arbitrary pair of dimensions valid. Resolutions above 2560 × 1440 are described as experimental. Keep important objects, text and useful negative space within the intended crop instead of relying on more pixels to resolve the composition.\n\nPNG is the default output format; JPEG and WebP are also supported. Compression from 0 to 100 can be selected for JPEG and WebP. Transparent, opaque and automatic backgrounds serve different layout needs. For transparency, choose PNG or WebP so the result can retain an alpha channel. A transparent asset can then be placed in another composition, while an opaque image supplies its own background. Keep the actual completed file together with the brief and selected output settings.\n\nOpenAI documents direct generation and editing as well as a conversational image-generation tool that can retain prior results and apply follow-up instructions. In that conversational workflow, the action can be automatic, forced generation or forced editing when an image is in context. Use the chosen image as the starting point for the next change rather than silently switching to another variation. The documented streaming workflow can deliver up to three partial images before the final result; a partial preview helps show progress but is not the finished asset. The manufacturer interface also supports batch processing.\n\nCheck the edited subject, lettering, object placement and the boundary of a masked change at the intended display size. GPT Image can still struggle with precise text positioning, structured layouts or consistency across recurring subjects and brand elements. A focused next instruction should identify the visible issue and the elements to preserve. Sunburst is useful for a precise, reference-led creative workflow whose final artifact can be assessed against a concrete visual brief, with higher quality, dimensions and transparency chosen for that artifact.", "editorialOverlay": { "copy_fields": null, "copy_sha256": null, "copy_source": null, "html_sha256": "a136535563188a15207ac13d85ec6ac26c209d9f66f6a813f7efa0814e7db8b2", "module_sha256": "ee9dc1c50b1e6a8fb7fa062030fcb4646b9aaaeb08be513b723a5a92a7290f06", "raw_model_unchanged": true, "schema": "ailoot-native-model-editorial-v1" }, "locale": "en", "methods": [ { "html": "openai/gpt-image-2.5-sunburstGPT Image 2.5 Sunburst generates and edits images from text and visual references, with an emphasis on editing precision. It supports masks, iterative refinement, custom frame sizes, transparent backgrounds and quality levels from low through max.
This description covers the model’s capabilities. Available in Ailoot inputs, outputs and options are listed for the selected method below.
GPT Image 2.5 Sunburst is OpenAI’s precision-oriented model for image generation and editing. It accepts text and image inputs and produces image outputs. Use it when a creative task depends on carefully modifying existing material, working with visual references or refining a composition through several steps. The brief can create a new scene, transform an existing image or guide an edit to a selected region. Sunburst and Flare are separate editions: Sunburst emphasises editing precision, while Flare is positioned for fast everyday generation.
For a new image, describe the subject, setting, composition, lighting and style, then state the intended shape and use. For a product concept, request a glass lamp on a stone surface with soft side lighting, a clearly visible silhouette and open space for a headline. For an edit, name the change and the elements that should remain recognisable: replace the background, adjust the light or alter a material while retaining the selected composition. The direct producer generation and edit workflows permit a prompt up to 32,000 characters and one to ten results.
Reference editing can use up to sixteen input images. Give each image a role so the instruction distinguishes the subject to retain, an environment to use and a style to apply. The model can also work with a mask that indicates an area to change. The source and mask need the same format and dimensions, and the mask must carry an alpha channel; the guide gives a size requirement below 50 MB. With several source images, the mask applies to the first. It guides the requested change rather than guaranteeing a hard boundary, so include preservation instructions in the brief and inspect the surrounding area in the result.
Quality settings are low, medium, high, xhigh, max and auto, with auto as the default. Low can support a quick visual draft, while higher settings are useful to compare when fine detail matters. Quality is separate from output dimensions and file compression. Standard sizes include 1024 × 1024, 1536 × 1024 and 1024 × 1536 pixels, and automatic or custom dimensions are also available. Choose the shape around the final layout before deciding how much detail the image needs.
Custom dimensions must be multiples of sixteen pixels, neither edge may exceed 3840 pixels, the aspect ratio must stay between 1:3 and 3:1, and total area must lie between 655,360 and 8,294,400 pixels. These conditions apply together. A 3840 × 2160 landscape fits the upper area limit, while a long edge by itself does not make an arbitrary pair of dimensions valid. Resolutions above 2560 × 1440 are described as experimental. Keep important objects, text and useful negative space within the intended crop instead of relying on more pixels to resolve the composition.
PNG is the default output format; JPEG and WebP are also supported. Compression from 0 to 100 can be selected for JPEG and WebP. Transparent, opaque and automatic backgrounds serve different layout needs. For transparency, choose PNG or WebP so the result can retain an alpha channel. A transparent asset can then be placed in another composition, while an opaque image supplies its own background. Keep the actual completed file together with the brief and selected output settings.
OpenAI documents direct generation and editing as well as a conversational image-generation tool that can retain prior results and apply follow-up instructions. In that conversational workflow, the action can be automatic, forced generation or forced editing when an image is in context. Use the chosen image as the starting point for the next change rather than silently switching to another variation. The documented streaming workflow can deliver up to three partial images before the final result; a partial preview helps show progress but is not the finished asset. The manufacturer interface also supports batch processing.
Check the edited subject, lettering, object placement and the boundary of a masked change at the intended display size. GPT Image can still struggle with precise text positioning, structured layouts or consistency across recurring subjects and brand elements. A focused next instruction should identify the visible issue and the elements to preserve. Sunburst is useful for a precise, reference-led creative workflow whose final artifact can be assessed against a concrete visual brief, with higher quality, dimensions and transparency chosen for that artifact.
POST /v1/images/generationstext, imageimage| Field | Type | Requirement | Values & limits | Default |
|---|---|---|---|---|
Fieldmodel | Typestring | RequirementRequired | Values & limits openai/gpt-image-2.5-sunburst | Default— |
Fieldprompt | Typestring | RequirementRequired | Values & limits Minimum length: 1 | Default— |
Fieldaspect_ratio | Typestring | RequirementOptional | Values & limits 1:1, 3:2, 2:3, 4:3, 3:4, 16:9, 9:16, 21:9, auto | Default— |
Fieldquality | Typestring | RequirementOptional | Values & limits auto, low, medium, high, xhigh, max | Default— |
Fieldbackground | Typestring | RequirementOptional | Values & limits auto, transparent, opaque | Default— |
Fieldn | Typeinteger | RequirementOptional | Values & limits Minimum: 1 Maximum: 10 | Default— |
Fieldinput_references | Typearray | RequirementOptional | Values & limits Minimum items: 0 Maximum items: 16 | Default— |
Fieldoutput_compression | Typeinteger | RequirementOptional | Values & limits Minimum: 0 Maximum: 100 | Default— |
Fieldprovider | Typeobject | RequirementOptional | Values & limits | Default— |
| Parameter | Contract |
|---|---|
Parametercontent_type | Contractapplication/json |
Parameterrequired | Contractmodel, prompt |
Parametermaterial_types | Contracttext, image |
Lifecycle: synchronous_result_then_owned_library
GET /v1/media/tasks/{task_id} · actual returned task_id or media library record
GET /v1/media/assets/{asset_id}/content
GET /v1/media/assets
GET /v1/media/assets/{asset_id}
The displayed rates estimate the cost before generation. The final charge is calculated from the cost recorded for the original generation. Measured output duration and forecast quantities are not the final billable quantity.
Tariff references do not add request options. Supported parameters are listed for the selected method.
Match each charge to the conditions shown. Use one rate for each set of alternative configurations.
| Configuration | Rate schedule | Forecast quantity | Charge |
|---|---|---|---|
| ConfigurationBase token rates | Rate scheduleBase rate | Forecast quantity— | Charge Input · Input tokens · Estimate for comparison only — / 1K tokens Output · Output tokens · Estimate for comparison only — / 1K tokens Input · Cache read · Estimate for comparison only — / 1K tokens Input · Cache write · Estimate for comparison only — / 1K tokens Input · Cache write, 1 hour · Estimate for comparison only — / 1K tokens |
| ConfigurationEstimated generation cost | Rate schedule— | Forecast quantityone generation estimate | Charge Output · Estimated generation cost · Estimate for comparison only $0.24576 estimate before generation This estimate is not an additional charge or a fixed rate for the completed generation. |
Ailoot reserves an estimated amount before a request. After the result, it settles the final charge from measured usage. Streaming settlement uses the final usage record. How reservation and settlement work →
set -eu\n: "${AILOOT_API_KEY:?Set your Ailoot API key}"\nexport AILOOT_API_KEY\npython3 - <<'PY_REQUEST'\nimport json,os\nfrom pathlib import Path\np={'model': 'openai/gpt-image-2.5-sunburst', 'prompt': 'a red panda astronaut floating in space, studio lighting'}\n\nPath("request.json").write_text(json.dumps(p,ensure_ascii=False))\nPY_REQUEST\ncurl --fail-with-body --silent --show-error https://api.ailoot.ai/v1/images/generations \\\n -H "Authorization: Bearer $AILOOT_API_KEY" -H 'Content-Type: application/json' \\\n --data-binary @request.json -D response-headers.txt -o result.json\npython3 - <<'PY_RESULT'\nimport base64, json, os, subprocess\nfrom pathlib import Path\nfrom urllib.parse import urlsplit, urljoin\n\ndef save_images(items):\n if not items: raise SystemExit("No completed images; retain the full response")\n for index, item in enumerate(items):\n encoded=item.get("b64_json")\n uri=item.get("url", "")\n if not encoded and uri.startswith("data:image/"):\n header, separator, encoded=uri.partition(",")\n if not separator or ";base64" not in header: raise SystemExit("Unsupported image data URI")\n if encoded:\n raw=base64.b64decode(encoded,validate=True)\n elif uri:\n uri=urljoin("https://api.ailoot.ai",uri)\n parsed=urlsplit(uri)\n if parsed.scheme!="https" or not parsed.netloc or parsed.username or parsed.password:\n raise SystemExit("Invalid returned image URL")\n owned=parsed.netloc=="api.ailoot.ai" and parsed.path.startswith("/v1/media/assets/") and parsed.path.endswith("/content")\n args=["curl","--fail-with-body","--silent","--show-error",uri]\n if owned: args += ["-H","Authorization: Bearer "+os.environ["AILOOT_API_KEY"]]\n temporary=Path(f"image-{index}.bin")\n subprocess.run(args+["-o",str(temporary)],check=True)\n raw=temporary.read_bytes()\n temporary.unlink()\n else: raise SystemExit("Image has neither bytes nor URL")\n if not raw: raise SystemExit("Empty image result")\n ext="png" if raw.startswith(b"\\x89PNG") else "jpg" if raw.startswith(b"\\xff\\xd8") else "webp" if raw[:4]==b"RIFF" and raw[8:12]==b"WEBP" else "svg" if raw.lstrip().startswith((b"<svg",b"<?xml")) else "bin"\n Path(f"image-{index}.{ext}").write_bytes(raw)\n print("Every returned image saved; complete response metadata retained")\nsave_images(json.loads(Path("result.json").read_text())["data"])\nPY_RESULT\nDownload each returned image or save its Base64 contents
GET /v1/media/tasks/{task_id}GET /v1/media/assets/{asset_id}/contentGET /v1/media/assetsGET /v1/media/assets/{asset_id}set -eu\n: "${AILOOT_API_KEY:?Set your Ailoot API key}"\ncurl --fail-with-body --silent --show-error https://api.ailoot.ai/v1/media/assets \\\n -H "Authorization: Bearer $AILOOT_API_KEY" -o owned-assets.json\n# Preserve every returned asset and follow its exact signed content URL. Do not construct a signed token.\nSave every completed image; retrieve each returned URL or decode its Base64 contents.
Retain the complete response and headers, every actual output, original returned identifiers and all usage fields. Recover the original operation without submitting creation again.
Keep the returned X-Client-Request-ID for reconciliation. Read only actual returned task_id/asset_id values or the owned media library; never build a task URL from the operation ID and never resubmit creation.
| Field | Type |
|---|---|
Fieldcreated | Type |
Fielddata[].url | Type |
Fielddata[].b64_json | Type |
Fieldusage | Type |
Ailoot Ultra openai/gpt-image-2.5-sunburst
/v1/images