FLUX.2 [pro] is Black Forest Labs’ production-focused image-generation and editing model. The manufacturer presents it as the default choice for most FLUX.2 visual tasks: create a new image from text or revise a scene using visual references. Its family capabilities include photorealistic materials and lighting, readable text, exact-color instructions and consistent subjects across related edits. The returned artifact is an image to use in a product scene, editorial composition, design concept or visual communication.
Begin with the subject, what it is doing or how it is arranged, the style, and the setting or light. The manufacturer’s prompt guide gives more attention to early words and recommends placing the essential subject and action before decorative details. It suggests 10–30 words for quick concepts and 30–80 for ordinary projects, with longer descriptions for complex scenes. Those ranges are prompt-writing guidance, not a text-context limit.
An original catalog brief might show a brushed-steel lamp on a warm-grey tabletop, with a three-quarter view that reveals the switch and shade. Specify the surface, illumination and viewpoint that make the product recognizable. For an infographic or marketing image, settle the visible wording and hierarchy first; quote the exact title, identify its placement and name the style and size of the lettering. Inspect labels, quantities and product details at the final display size.
The family guide documents both natural-language and JSON-structured prompts. A structured scene brief can keep the subject, background, light and composition in separate fields so one requirement can be revised clearly. It also supports hex-color descriptions: associate the code with the specific object or text to color. Describe the desired appearance positively, because the documented FLUX.2 prompting flow does not provide negative prompts.
The exact [pro] request profile offers eight reference-image positions; the maker’s interactive interface separately supports ten. Assign each picture a role, such as subject appearance, clothing, product design, room or style. For an existing scene, state the element to change and which composition, surroundings and visual details should remain. The published examples include furniture composites, background or weather changes, replacement objects and text revisions.
Reference capacity also depends on output size. The producer’s [pro] guide specifies a 9-megapixel total budget for input plus output. Its examples allow eight references at 1-megapixel output and seven at 2-megapixel output, reducing reference capacity as the requested result grows. Plan the source set and output together instead of treating the maximum slot count as available for every output size. Use the references that establish necessary information and identify their intended contribution.
For a product composite, the first image could supply the product and the second a room, while a third establishes a material finish. Explain how they should become one coherent scene. If the result changes a shape, marking or identity meant to remain, target that issue in the follow-up rather than replacing the entire brief. The practical result is the revised picture with reference fidelity and the specific change compared against its inputs.
The exact [pro] specification enables prompt upsampling by default. Its disable_pup control stops that expansion when a carefully written instruction should be used as supplied. Retain the chosen setting when comparing versions, because the same short concept and an expanded description are different generation inputs. This is a maker-profile control, rather than a guarantee that every requested detail appears correctly.
The documentation distinguishes the plain [pro] endpoint as a fixed snapshot from [pro] Preview, where subsequent model improvements are introduced. Their request and response shapes are the same, while their underlying weights differ. Keep the exact selected profile with the image and references. The separate Preview label should not silently replace a plain [pro] selection when a revision depends on the same documented model choice.
The family guide documents output up to 4 megapixels, with explicit dimensions at least 64×64 and multiples of 16. It recommends up to 2 megapixels for most use cases. Width and height establish the intended framing; a square product image, portrait layout and wide editorial scene distribute space differently. The exact request lists JPEG, PNG and WebP output, with JPEG as the default, and an optional seed.
The [max] profile has a separate grounding-search feature, while [flex] exposes sampling steps and guidance and [klein] has a smaller, four-reference profile. Select the edition around the required work rather than importing those controls into [pro]. For the selected asset, retain the actual image, final prompt, references, dimensions and model label. The producer’s FLUX.2 signed result links expire after ten minutes, so retrieve the image instead of relying on its temporary location. Review subject identity, colors, text, composition and the edited details before placing the result in the customer-facing material.
", "aboutText": "FLUX.2 [pro] is Black Forest Labs’ production-focused image-generation and editing model. The manufacturer presents it as the default choice for most FLUX.2 visual tasks: create a new image from text or revise a scene using visual references. Its family capabilities include photorealistic materials and lighting, readable text, exact-color instructions and consistent subjects across related edits. The returned artifact is an image to use in a product scene, editorial composition, design concept or visual communication.\n\n### Prepare a production brief\n\nBegin with the subject, what it is doing or how it is arranged, the style, and the setting or light. The manufacturer’s prompt guide gives more attention to early words and recommends placing the essential subject and action before decorative details. It suggests 10–30 words for quick concepts and 30–80 for ordinary projects, with longer descriptions for complex scenes. Those ranges are prompt-writing guidance, not a text-context limit.\n\nAn original catalog brief might show a brushed-steel lamp on a warm-grey tabletop, with a three-quarter view that reveals the switch and shade. Specify the surface, illumination and viewpoint that make the product recognizable. For an infographic or marketing image, settle the visible wording and hierarchy first; quote the exact title, identify its placement and name the style and size of the lettering. Inspect labels, quantities and product details at the final display size.\n\nThe family guide documents both natural-language and JSON-structured prompts. A structured scene brief can keep the subject, background, light and composition in separate fields so one requirement can be revised clearly. It also supports hex-color descriptions: associate the code with the specific object or text to color. Describe the desired appearance positively, because the documented FLUX.2 prompting flow does not provide negative prompts.\n\n### Edit with references and a shared pixel budget\n\nThe exact [pro] request profile offers eight reference-image positions; the maker’s interactive interface separately supports ten. Assign each picture a role, such as subject appearance, clothing, product design, room or style. For an existing scene, state the element to change and which composition, surroundings and visual details should remain. The published examples include furniture composites, background or weather changes, replacement objects and text revisions.\n\nReference capacity also depends on output size. The producer’s [pro] guide specifies a 9-megapixel total budget for input plus output. Its examples allow eight references at 1-megapixel output and seven at 2-megapixel output, reducing reference capacity as the requested result grows. Plan the source set and output together instead of treating the maximum slot count as available for every output size. Use the references that establish necessary information and identify their intended contribution.\n\nFor a product composite, the first image could supply the product and the second a room, while a third establishes a material finish. Explain how they should become one coherent scene. If the result changes a shape, marking or identity meant to remain, target that issue in the follow-up rather than replacing the entire brief. The practical result is the revised picture with reference fidelity and the specific change compared against its inputs.\n\n### Prompt expansion and fixed versus preview profiles\n\nThe exact [pro] specification enables prompt upsampling by default. Its disable_pup control stops that expansion when a carefully written instruction should be used as supplied. Retain the chosen setting when comparing versions, because the same short concept and an expanded description are different generation inputs. This is a maker-profile control, rather than a guarantee that every requested detail appears correctly.\n\nThe documentation distinguishes the plain [pro] endpoint as a fixed snapshot from [pro] Preview, where subsequent model improvements are introduced. Their request and response shapes are the same, while their underlying weights differ. Keep the exact selected profile with the image and references. The separate Preview label should not silently replace a plain [pro] selection when a revision depends on the same documented model choice.\n\n### Dimensions, formats and the selected asset\n\nThe family guide documents output up to 4 megapixels, with explicit dimensions at least 64×64 and multiples of 16. It recommends up to 2 megapixels for most use cases. Width and height establish the intended framing; a square product image, portrait layout and wide editorial scene distribute space differently. The exact request lists JPEG, PNG and WebP output, with JPEG as the default, and an optional seed.\n\nThe [max] profile has a separate grounding-search feature, while [flex] exposes sampling steps and guidance and [klein] has a smaller, four-reference profile. Select the edition around the required work rather than importing those controls into [pro]. For the selected asset, retain the actual image, final prompt, references, dimensions and model label. The producer’s FLUX.2 signed result links expire after ten minutes, so retrieve the image instead of relying on its temporary location. Review subject identity, colors, text, composition and the edited details before placing the result in the customer-facing material.", "editorialOverlay": { "copy_fields": null, "copy_sha256": null, "copy_source": null, "html_sha256": "7da8c56a7f56193d7560dcd8acefa04e1c03eb5ac83e3f52d30cff6c9153b946", "module_sha256": "ee9dc1c50b1e6a8fb7fa062030fcb4646b9aaaeb08be513b723a5a92a7290f06", "raw_model_unchanged": true, "schema": "ailoot-native-model-editorial-v1" }, "locale": "en", "methods": [ { "html": "black-forest-labs/flux.2-proFLUX.2 [pro] is Black Forest Labs’ production image model for text-led creation and reference-guided editing. Its documented profile combines photorealistic detail, text and color control with a reference/output pixel budget and up to 4-megapixel images.
This description covers the model’s capabilities. Available in Ailoot inputs, outputs and options are listed for the selected method below.
FLUX.2 [pro] is Black Forest Labs’ production-focused image-generation and editing model. The manufacturer presents it as the default choice for most FLUX.2 visual tasks: create a new image from text or revise a scene using visual references. Its family capabilities include photorealistic materials and lighting, readable text, exact-color instructions and consistent subjects across related edits. The returned artifact is an image to use in a product scene, editorial composition, design concept or visual communication.
Begin with the subject, what it is doing or how it is arranged, the style, and the setting or light. The manufacturer’s prompt guide gives more attention to early words and recommends placing the essential subject and action before decorative details. It suggests 10–30 words for quick concepts and 30–80 for ordinary projects, with longer descriptions for complex scenes. Those ranges are prompt-writing guidance, not a text-context limit.
An original catalog brief might show a brushed-steel lamp on a warm-grey tabletop, with a three-quarter view that reveals the switch and shade. Specify the surface, illumination and viewpoint that make the product recognizable. For an infographic or marketing image, settle the visible wording and hierarchy first; quote the exact title, identify its placement and name the style and size of the lettering. Inspect labels, quantities and product details at the final display size.
The family guide documents both natural-language and JSON-structured prompts. A structured scene brief can keep the subject, background, light and composition in separate fields so one requirement can be revised clearly. It also supports hex-color descriptions: associate the code with the specific object or text to color. Describe the desired appearance positively, because the documented FLUX.2 prompting flow does not provide negative prompts.
The exact [pro] request profile offers eight reference-image positions; the maker’s interactive interface separately supports ten. Assign each picture a role, such as subject appearance, clothing, product design, room or style. For an existing scene, state the element to change and which composition, surroundings and visual details should remain. The published examples include furniture composites, background or weather changes, replacement objects and text revisions.
Reference capacity also depends on output size. The producer’s [pro] guide specifies a 9-megapixel total budget for input plus output. Its examples allow eight references at 1-megapixel output and seven at 2-megapixel output, reducing reference capacity as the requested result grows. Plan the source set and output together instead of treating the maximum slot count as available for every output size. Use the references that establish necessary information and identify their intended contribution.
For a product composite, the first image could supply the product and the second a room, while a third establishes a material finish. Explain how they should become one coherent scene. If the result changes a shape, marking or identity meant to remain, target that issue in the follow-up rather than replacing the entire brief. The practical result is the revised picture with reference fidelity and the specific change compared against its inputs.
The exact [pro] specification enables prompt upsampling by default. Its disable_pup control stops that expansion when a carefully written instruction should be used as supplied. Retain the chosen setting when comparing versions, because the same short concept and an expanded description are different generation inputs. This is a maker-profile control, rather than a guarantee that every requested detail appears correctly.
The documentation distinguishes the plain [pro] endpoint as a fixed snapshot from [pro] Preview, where subsequent model improvements are introduced. Their request and response shapes are the same, while their underlying weights differ. Keep the exact selected profile with the image and references. The separate Preview label should not silently replace a plain [pro] selection when a revision depends on the same documented model choice.
The family guide documents output up to 4 megapixels, with explicit dimensions at least 64×64 and multiples of 16. It recommends up to 2 megapixels for most use cases. Width and height establish the intended framing; a square product image, portrait layout and wide editorial scene distribute space differently. The exact request lists JPEG, PNG and WebP output, with JPEG as the default, and an optional seed.
The [max] profile has a separate grounding-search feature, while [flex] exposes sampling steps and guidance and [klein] has a smaller, four-reference profile. Select the edition around the required work rather than importing those controls into [pro]. For the selected asset, retain the actual image, final prompt, references, dimensions and model label. The producer’s FLUX.2 signed result links expire after ten minutes, so retrieve the image instead of relying on its temporary location. Review subject identity, colors, text, composition and the edited details before placing the result in the customer-facing material.
POST /v1/images/generationstext, imageimage| Field | Type | Requirement | Values & limits | Default |
|---|---|---|---|---|
Fieldmodel | Typestring | RequirementRequired | Values & limits black-forest-labs/flux.2-pro | Default— |
Fieldprompt | Typestring | RequirementRequired | Values & limits Minimum length: 1 | Default— |
Fieldaspect_ratio | Typestring | RequirementOptional | Values & limits 1:1, 4:3, 3:4, 3:2, 2:3, 16:9, 9:16, 21:9, auto | Default— |
Fieldoutput_format | Typestring | RequirementOptional | Values & limits png, jpeg | Default— |
Fieldn | Typeinteger | RequirementOptional | Values & limits Minimum: 1 Maximum: 1 | Default— |
Fieldinput_references | Typearray | RequirementOptional | Values & limits Minimum items: 0 Maximum items: 8 | 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 | Billing unit | Charge |
|---|---|---|---|---|
| ConfigurationBase token rates | Rate scheduleBase rate | Forecast quantity— | Billing unit— | 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 |
| Configurationoutput image megapixel | Rate schedule— | Forecast quantity— | Billing unitper output image megapixel | Charge Output · output image megapixel · Estimate for comparison only $0.06 per output image megapixel |
| ConfigurationEstimated generation cost | Rate schedule— | Forecast quantityone generation estimate | Billing unit— | Charge Output · Estimated generation cost · Estimate for comparison only $0.06 estimate before generation This estimate is not an additional charge or a fixed rate for the completed generation. |
Ailoot reserves an estimated amount before a request. After the result, it settles the final charge from measured usage. Streaming settlement uses the final usage record. How reservation and settlement work →
set -eu\n: "${AILOOT_API_KEY:?Set your Ailoot API key}"\nexport AILOOT_API_KEY\npython3 - <<'PY_REQUEST'\nimport json,os\nfrom pathlib import Path\np={'model': 'black-forest-labs/flux.2-pro', '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 black-forest-labs/flux.2-pro
/v1/images