Microsoft AI MAI-Image-2.6 Flash is the optimized member of the 2.6 image family. It performs text-to-image generation and image-to-image editing, with the same reference, grounding and composition controls documented for MAI-Image-2.6. Microsoft positions Flash for work that needs a responsive sequence of images or edits, such as visual concepts, repeated content production and revisions of an existing design. The model produces a completed image rather than a design project with independently editable objects.
For a new image, prepare the subject, intended arrangement, lighting, material cues and style in the brief. Specify exact wording and placement for any visible text. The 2.6 family’s documented strengths include text rendering, portraits, photorealistic commercial visuals and 3D-style imagery. A useful production brief makes clear which details should stay consistent across versions and which may vary. For example, a series of concept images can retain the same product description and colour palette while changing the surrounding scene or camera viewpoint.
For edits, give the source image and name the intended correction. The manufacturer describes targeted object edits, layout adaptation, text changes and cleanup of artifacts such as motion blur. The important preparation is a clear relationship between the change and the retained composition. For a stationery edit, replace the surface beneath the set while preserving the arrangement, print on the notebook and direction of the shadows. Review the complete result for unintended changes to those retained elements, then make the next requested correction specific.
Multi-reference editing accepts up to five JPEG or PNG images in the manufacturer’s image interface. Use separate references for the object, scene or style where those roles help explain the desired result. Give each reference a short role in the brief instead of leaving the model to choose which image supplies the product or the layout. This is useful when preparing a consistent campaign concept: one reference identifies the item, another the intended environment and a third the visual treatment. The generated image still needs inspection for likeness, scale and text accuracy.
Optional web grounding can add relevant current context to the 2.6 generation or editing request. Use a precise subject description when the desired image depends on an actual place or an identifiable object. Automatic aspect selection can choose a shape from the brief and supplied references, while explicit sizing remains useful when a layout has a known destination. Prepare the intended crop and the area reserved for later text before asking for variants. Changing a crop after composition can remove the visual detail that made the image useful.
The output-area limit is 2,359,296 pixels, equivalent to a 1536×1536 square. The serving guide requires both image dimensions to be at least 768 pixels and allows either side to exceed 1536 when the total area stays within the limit. A landscape or portrait target therefore needs a width-and-height calculation rather than an assumed maximum on each side. The manufacturer’s generation and editing responses contain PNG image data. Save the complete picture for inspection at the size where it will appear, including small lettering and the edges of the main object.
The joint 2.6 model card reports a 32K-token context. The serving guide specifies a 32,000-token maximum prompt context, and the exact Flash catalogue entry lists 4,096 output tokens. Those token quantities are separate from the image’s pixel area. A concise, well-organized brief can still be preferable to a long collection of contradictory descriptions: separate the subject, retained details, requested change and visual treatment. Keep the source references and final wording with the image so a later revision starts from the actual design intent.
The precision edition MAI-Image-2.6 and this Flash edition share the 2.6 creation controls. Flash’s distinguishing role is efficient, responsive production; selecting it does not add a fixed generation-time guarantee. The older 2.5 family has a smaller documented output-area allowance and its own edition choices. The 2.6 grounding and model-directed aspect controls are specific to the 2.6 family. This distinction helps preserve the same preparation workflow where it is documented while keeping version-specific settings separate.
For a set of generated images, compare repeated objects, faces, lettering, palette and composition before arranging them as one design system. For an edit, compare the result with the actual source rather than judging only the changed object. Photorealistic detail can be persuasive while a small label, reflection or geometric relationship is still incorrect. A practical review identifies that specific defect and retains the other successful choices for the next revision.
", "aboutText": "Microsoft AI MAI-Image-2.6 Flash is the optimized member of the 2.6 image family. It performs text-to-image generation and image-to-image editing, with the same reference, grounding and composition controls documented for MAI-Image-2.6. Microsoft positions Flash for work that needs a responsive sequence of images or edits, such as visual concepts, repeated content production and revisions of an existing design. The model produces a completed image rather than a design project with independently editable objects.\n\nFor a new image, prepare the subject, intended arrangement, lighting, material cues and style in the brief. Specify exact wording and placement for any visible text. The 2.6 family’s documented strengths include text rendering, portraits, photorealistic commercial visuals and 3D-style imagery. A useful production brief makes clear which details should stay consistent across versions and which may vary. For example, a series of concept images can retain the same product description and colour palette while changing the surrounding scene or camera viewpoint.\n\nFor edits, give the source image and name the intended correction. The manufacturer describes targeted object edits, layout adaptation, text changes and cleanup of artifacts such as motion blur. The important preparation is a clear relationship between the change and the retained composition. For a stationery edit, replace the surface beneath the set while preserving the arrangement, print on the notebook and direction of the shadows. Review the complete result for unintended changes to those retained elements, then make the next requested correction specific.\n\nMulti-reference editing accepts up to five JPEG or PNG images in the manufacturer’s image interface. Use separate references for the object, scene or style where those roles help explain the desired result. Give each reference a short role in the brief instead of leaving the model to choose which image supplies the product or the layout. This is useful when preparing a consistent campaign concept: one reference identifies the item, another the intended environment and a third the visual treatment. The generated image still needs inspection for likeness, scale and text accuracy.\n\nOptional web grounding can add relevant current context to the 2.6 generation or editing request. Use a precise subject description when the desired image depends on an actual place or an identifiable object. Automatic aspect selection can choose a shape from the brief and supplied references, while explicit sizing remains useful when a layout has a known destination. Prepare the intended crop and the area reserved for later text before asking for variants. Changing a crop after composition can remove the visual detail that made the image useful.\n\nThe output-area limit is 2,359,296 pixels, equivalent to a 1536×1536 square. The serving guide requires both image dimensions to be at least 768 pixels and allows either side to exceed 1536 when the total area stays within the limit. A landscape or portrait target therefore needs a width-and-height calculation rather than an assumed maximum on each side. The manufacturer’s generation and editing responses contain PNG image data. Save the complete picture for inspection at the size where it will appear, including small lettering and the edges of the main object.\n\nThe joint 2.6 model card reports a 32K-token context. The serving guide specifies a 32,000-token maximum prompt context, and the exact Flash catalogue entry lists 4,096 output tokens. Those token quantities are separate from the image’s pixel area. A concise, well-organized brief can still be preferable to a long collection of contradictory descriptions: separate the subject, retained details, requested change and visual treatment. Keep the source references and final wording with the image so a later revision starts from the actual design intent.\n\nThe precision edition MAI-Image-2.6 and this Flash edition share the 2.6 creation controls. Flash’s distinguishing role is efficient, responsive production; selecting it does not add a fixed generation-time guarantee. The older 2.5 family has a smaller documented output-area allowance and its own edition choices. The 2.6 grounding and model-directed aspect controls are specific to the 2.6 family. This distinction helps preserve the same preparation workflow where it is documented while keeping version-specific settings separate.\n\nFor a set of generated images, compare repeated objects, faces, lettering, palette and composition before arranging them as one design system. For an edit, compare the result with the actual source rather than judging only the changed object. Photorealistic detail can be persuasive while a small label, reflection or geometric relationship is still incorrect. A practical review identifies that specific defect and retains the other successful choices for the next revision.", "editorialOverlay": { "copy_fields": null, "copy_sha256": null, "copy_source": null, "html_sha256": "3518e94b9317eeba29a79be03b3cab9b2b8d5246a8e633630e1b83e7fc8fad37", "module_sha256": "ee9dc1c50b1e6a8fb7fa062030fcb4646b9aaaeb08be513b723a5a92a7290f06", "raw_model_unchanged": true, "schema": "ailoot-native-model-editorial-v1" }, "locale": "en", "methods": [ { "html": "microsoft/mai-image-2.6-flashMicrosoft AI MAI-Image-2.6 Flash generates and edits PNG images for repeated creative production. It retains multi-image references, optional web grounding and automatic aspect selection, with up to 2,359,296 output pixels. Flash is the 2.6 edition optimized for responsive generation and editing workflows.
This description covers the model’s capabilities. Available in Ailoot inputs, outputs and options are listed for the selected method below.
Microsoft AI MAI-Image-2.6 Flash is the optimized member of the 2.6 image family. It performs text-to-image generation and image-to-image editing, with the same reference, grounding and composition controls documented for MAI-Image-2.6. Microsoft positions Flash for work that needs a responsive sequence of images or edits, such as visual concepts, repeated content production and revisions of an existing design. The model produces a completed image rather than a design project with independently editable objects.
For a new image, prepare the subject, intended arrangement, lighting, material cues and style in the brief. Specify exact wording and placement for any visible text. The 2.6 family’s documented strengths include text rendering, portraits, photorealistic commercial visuals and 3D-style imagery. A useful production brief makes clear which details should stay consistent across versions and which may vary. For example, a series of concept images can retain the same product description and colour palette while changing the surrounding scene or camera viewpoint.
For edits, give the source image and name the intended correction. The manufacturer describes targeted object edits, layout adaptation, text changes and cleanup of artifacts such as motion blur. The important preparation is a clear relationship between the change and the retained composition. For a stationery edit, replace the surface beneath the set while preserving the arrangement, print on the notebook and direction of the shadows. Review the complete result for unintended changes to those retained elements, then make the next requested correction specific.
Multi-reference editing accepts up to five JPEG or PNG images in the manufacturer’s image interface. Use separate references for the object, scene or style where those roles help explain the desired result. Give each reference a short role in the brief instead of leaving the model to choose which image supplies the product or the layout. This is useful when preparing a consistent campaign concept: one reference identifies the item, another the intended environment and a third the visual treatment. The generated image still needs inspection for likeness, scale and text accuracy.
Optional web grounding can add relevant current context to the 2.6 generation or editing request. Use a precise subject description when the desired image depends on an actual place or an identifiable object. Automatic aspect selection can choose a shape from the brief and supplied references, while explicit sizing remains useful when a layout has a known destination. Prepare the intended crop and the area reserved for later text before asking for variants. Changing a crop after composition can remove the visual detail that made the image useful.
The output-area limit is 2,359,296 pixels, equivalent to a 1536×1536 square. The serving guide requires both image dimensions to be at least 768 pixels and allows either side to exceed 1536 when the total area stays within the limit. A landscape or portrait target therefore needs a width-and-height calculation rather than an assumed maximum on each side. The manufacturer’s generation and editing responses contain PNG image data. Save the complete picture for inspection at the size where it will appear, including small lettering and the edges of the main object.
The joint 2.6 model card reports a 32K-token context. The serving guide specifies a 32,000-token maximum prompt context, and the exact Flash catalogue entry lists 4,096 output tokens. Those token quantities are separate from the image’s pixel area. A concise, well-organized brief can still be preferable to a long collection of contradictory descriptions: separate the subject, retained details, requested change and visual treatment. Keep the source references and final wording with the image so a later revision starts from the actual design intent.
The precision edition MAI-Image-2.6 and this Flash edition share the 2.6 creation controls. Flash’s distinguishing role is efficient, responsive production; selecting it does not add a fixed generation-time guarantee. The older 2.5 family has a smaller documented output-area allowance and its own edition choices. The 2.6 grounding and model-directed aspect controls are specific to the 2.6 family. This distinction helps preserve the same preparation workflow where it is documented while keeping version-specific settings separate.
For a set of generated images, compare repeated objects, faces, lettering, palette and composition before arranging them as one design system. For an edit, compare the result with the actual source rather than judging only the changed object. Photorealistic detail can be persuasive while a small label, reflection or geometric relationship is still incorrect. A practical review identifies that specific defect and retains the other successful choices for the next revision.
POST /v1/images/generationstext, imageimage| Field | Type | Requirement | Values & limits | Default |
|---|---|---|---|---|
Fieldmodel | Typestring | RequirementRequired | Values & limits microsoft/mai-image-2.6-flash | Default— |
Fieldprompt | Typestring | RequirementRequired | Values & limits Minimum length: 1 | Default— |
Fieldaspect_ratio | Typestring | RequirementOptional | Values & limits 1:1, 4:3, 3:4, 16:9, 9:16, 3:2, 2:3, auto | Default— |
Fieldn | Typeinteger | RequirementOptional | Values & limits Minimum: 1 Maximum: 1 | Default— |
Fieldinput_references | Typearray | RequirementOptional | Values & limits Minimum items: 0 Maximum items: 5 | 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 $0.155648 estimate before generation This estimate is not an additional charge or a fixed rate for the completed generation. |
Ailoot reserves an estimated amount before a request. After the result, it settles the final charge from measured usage. Streaming settlement uses the final usage record. How reservation and settlement work →
set -eu\n: "${AILOOT_API_KEY:?Set your Ailoot API key}"\nexport AILOOT_API_KEY\npython3 - <<'PY_REQUEST'\nimport json,os\nfrom pathlib import Path\np={'model': 'microsoft/mai-image-2.6-flash', '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 microsoft/mai-image-2.6-flash
/v1/images