Qwen-Image-3.0 Pro is Qwen’s flagship third-generation model for image generation and editing. It combines complex layout creation, detailed realistic textures, multilingual typography and close following of an image brief. Pro and the standard Qwen-Image-3.0 share the documented generation and editing modes; Pro emphasises image quality, while the standard edition balances quality and speed. The result is a rendered image artifact.
For an information-rich graphic, define the content before the look: the headline, supporting text, figures, hierarchy and relationship between sections. A useful product brief can combine a main object, close-up material details and labelled features on a single canvas. For an educational illustration, identify the concept, the diagram structure and the exact captions instead of relying on the model to invent the underlying facts. Complex layouts can also show several panels or nested interfaces. Keep the recommended 4,500-token prompt maximum focused on what must appear and where it belongs.
The 3.0 family supports native typography in 12 languages, multiple fonts and more than 100 artistic styles, along with fine textures such as hair, skin or fabric. Small-text rendering is demonstrated down to 10 pixels. This supports a design that blends lettering with a physical material or combines text with realistic imagery. Give the precise wording, numerals and language, and describe the typographic role of each block. A newspaper page or interface mockup is still a picture, so any later editable layout or interactive behavior is a separate artifact to build from it.
Pro can edit text within an image, add, delete or move elements, change a subject’s action, transfer visual style and enhance details. Editing accepts one to three input images with an instruction. Their order determines image 1, image 2 and image 3 in the brief. Specify which reference supplies the subject’s identity, which provides an object or clothing and which contributes a setting or pose. A focused edit names both the intended change and the features that should stay intact, making the resulting image easier to compare with the references.
Supported inputs are JPG, JPEG, PNG, BMP, TIFF, WebP and GIF, with only the first frame used for animated GIFs. Each input image can be up to 10 MB. The exact API accepts sides from 1 to 8,000 pixels and a long-to-short-side ratio no greater than 10:1; its recommended range is 384–2,048 pixels. The editing guide separately recommends images up to 3,072 pixels, so treat the input choice as preparation for the detail needed rather than a promise that every resolution improves the result. Use a legible, adequately framed reference for tiny text or material features.
Outputs are PNG. The producer workflow returns one to six alternatives, defaulting to one, and supports a total pixel area from 512 × 512 to 2,048 × 2,048 with aspect ratios from 1:8 to 8:1. The pixel-area budget permits some non-square dimensions whose longer side exceeds 2,048 pixels. When no size is specified, the model recommends a resolution based on the brief. Choose the shape for the actual placement, then compare the generated alternatives for content completeness, subject details and typography at the intended display size.
Intelligent prompt rewriting is enabled by default. Direct enhancement is available for both generation and editing, while agent enhancement refines text-to-image prompts and cannot be used for image-to-image editing. Thinking is also enabled by default and requires prompt rewriting; it adds reasoning work and generation time. Negative prompts identify unwanted elements, and an optional seed can help make comparisons between variants more consistent. A clear positive brief should still establish the subject, composition and details to preserve.
A practical result-review step checks the exact text, numbers, reading order, reference fidelity and edges or textures that matter to the design. When a specific part needs revision, retain the successful composition and describe the focused change. Keep all selected images with their prompt and references, since producer output locations expire after 24 hours. Pro’s documented modes and limits belong to the 3.0 edition; the ten-reference and native transparency features described for Qwen 2.1 are separate edition capabilities.
", "aboutText": "Qwen-Image-3.0 Pro is Qwen’s flagship third-generation model for image generation and editing. It combines complex layout creation, detailed realistic textures, multilingual typography and close following of an image brief. Pro and the standard Qwen-Image-3.0 share the documented generation and editing modes; Pro emphasises image quality, while the standard edition balances quality and speed. The result is a rendered image artifact.\n\nFor an information-rich graphic, define the content before the look: the headline, supporting text, figures, hierarchy and relationship between sections. A useful product brief can combine a main object, close-up material details and labelled features on a single canvas. For an educational illustration, identify the concept, the diagram structure and the exact captions instead of relying on the model to invent the underlying facts. Complex layouts can also show several panels or nested interfaces. Keep the recommended 4,500-token prompt maximum focused on what must appear and where it belongs.\n\nThe 3.0 family supports native typography in 12 languages, multiple fonts and more than 100 artistic styles, along with fine textures such as hair, skin or fabric. Small-text rendering is demonstrated down to 10 pixels. This supports a design that blends lettering with a physical material or combines text with realistic imagery. Give the precise wording, numerals and language, and describe the typographic role of each block. A newspaper page or interface mockup is still a picture, so any later editable layout or interactive behavior is a separate artifact to build from it.\n\nPro can edit text within an image, add, delete or move elements, change a subject’s action, transfer visual style and enhance details. Editing accepts one to three input images with an instruction. Their order determines image 1, image 2 and image 3 in the brief. Specify which reference supplies the subject’s identity, which provides an object or clothing and which contributes a setting or pose. A focused edit names both the intended change and the features that should stay intact, making the resulting image easier to compare with the references.\n\nSupported inputs are JPG, JPEG, PNG, BMP, TIFF, WebP and GIF, with only the first frame used for animated GIFs. Each input image can be up to 10 MB. The exact API accepts sides from 1 to 8,000 pixels and a long-to-short-side ratio no greater than 10:1; its recommended range is 384–2,048 pixels. The editing guide separately recommends images up to 3,072 pixels, so treat the input choice as preparation for the detail needed rather than a promise that every resolution improves the result. Use a legible, adequately framed reference for tiny text or material features.\n\nOutputs are PNG. The producer workflow returns one to six alternatives, defaulting to one, and supports a total pixel area from 512 × 512 to 2,048 × 2,048 with aspect ratios from 1:8 to 8:1. The pixel-area budget permits some non-square dimensions whose longer side exceeds 2,048 pixels. When no size is specified, the model recommends a resolution based on the brief. Choose the shape for the actual placement, then compare the generated alternatives for content completeness, subject details and typography at the intended display size.\n\nIntelligent prompt rewriting is enabled by default. Direct enhancement is available for both generation and editing, while agent enhancement refines text-to-image prompts and cannot be used for image-to-image editing. Thinking is also enabled by default and requires prompt rewriting; it adds reasoning work and generation time. Negative prompts identify unwanted elements, and an optional seed can help make comparisons between variants more consistent. A clear positive brief should still establish the subject, composition and details to preserve.\n\nA practical result-review step checks the exact text, numbers, reading order, reference fidelity and edges or textures that matter to the design. When a specific part needs revision, retain the successful composition and describe the focused change. Keep all selected images with their prompt and references, since producer output locations expire after 24 hours. Pro’s documented modes and limits belong to the 3.0 edition; the ten-reference and native transparency features described for Qwen 2.1 are separate edition capabilities.", "editorialOverlay": { "copy_fields": null, "copy_sha256": null, "copy_source": null, "html_sha256": "90fd890d26abf09ebc06d99a63cdec4f06e40cb14fffb74077b5558e06e2d072", "module_sha256": "ee9dc1c50b1e6a8fb7fa062030fcb4646b9aaaeb08be513b723a5a92a7290f06", "raw_model_unchanged": true, "schema": "ailoot-native-model-editorial-v1" }, "locale": "en", "methods": [ { "html": "qwen/qwen-image-3-proQwen-Image-3.0 Pro is Qwen’s flagship image generation and editing model for complex layouts, realistic detail and multilingual text. Use a precise brief or one to three references to create a graphic, product scene or focused visual edit.
This description covers the model’s capabilities. Available in Ailoot inputs, outputs and options are listed for the selected method below.
Qwen-Image-3.0 Pro is Qwen’s flagship third-generation model for image generation and editing. It combines complex layout creation, detailed realistic textures, multilingual typography and close following of an image brief. Pro and the standard Qwen-Image-3.0 share the documented generation and editing modes; Pro emphasises image quality, while the standard edition balances quality and speed. The result is a rendered image artifact.
For an information-rich graphic, define the content before the look: the headline, supporting text, figures, hierarchy and relationship between sections. A useful product brief can combine a main object, close-up material details and labelled features on a single canvas. For an educational illustration, identify the concept, the diagram structure and the exact captions instead of relying on the model to invent the underlying facts. Complex layouts can also show several panels or nested interfaces. Keep the recommended 4,500-token prompt maximum focused on what must appear and where it belongs.
The 3.0 family supports native typography in 12 languages, multiple fonts and more than 100 artistic styles, along with fine textures such as hair, skin or fabric. Small-text rendering is demonstrated down to 10 pixels. This supports a design that blends lettering with a physical material or combines text with realistic imagery. Give the precise wording, numerals and language, and describe the typographic role of each block. A newspaper page or interface mockup is still a picture, so any later editable layout or interactive behavior is a separate artifact to build from it.
Pro can edit text within an image, add, delete or move elements, change a subject’s action, transfer visual style and enhance details. Editing accepts one to three input images with an instruction. Their order determines image 1, image 2 and image 3 in the brief. Specify which reference supplies the subject’s identity, which provides an object or clothing and which contributes a setting or pose. A focused edit names both the intended change and the features that should stay intact, making the resulting image easier to compare with the references.
Supported inputs are JPG, JPEG, PNG, BMP, TIFF, WebP and GIF, with only the first frame used for animated GIFs. Each input image can be up to 10 MB. The exact API accepts sides from 1 to 8,000 pixels and a long-to-short-side ratio no greater than 10:1; its recommended range is 384–2,048 pixels. The editing guide separately recommends images up to 3,072 pixels, so treat the input choice as preparation for the detail needed rather than a promise that every resolution improves the result. Use a legible, adequately framed reference for tiny text or material features.
Outputs are PNG. The producer workflow returns one to six alternatives, defaulting to one, and supports a total pixel area from 512 × 512 to 2,048 × 2,048 with aspect ratios from 1:8 to 8:1. The pixel-area budget permits some non-square dimensions whose longer side exceeds 2,048 pixels. When no size is specified, the model recommends a resolution based on the brief. Choose the shape for the actual placement, then compare the generated alternatives for content completeness, subject details and typography at the intended display size.
Intelligent prompt rewriting is enabled by default. Direct enhancement is available for both generation and editing, while agent enhancement refines text-to-image prompts and cannot be used for image-to-image editing. Thinking is also enabled by default and requires prompt rewriting; it adds reasoning work and generation time. Negative prompts identify unwanted elements, and an optional seed can help make comparisons between variants more consistent. A clear positive brief should still establish the subject, composition and details to preserve.
A practical result-review step checks the exact text, numbers, reading order, reference fidelity and edges or textures that matter to the design. When a specific part needs revision, retain the successful composition and describe the focused change. Keep all selected images with their prompt and references, since producer output locations expire after 24 hours. Pro’s documented modes and limits belong to the 3.0 edition; the ten-reference and native transparency features described for Qwen 2.1 are separate edition capabilities.
POST /v1/images/generationstext, imageimage| Field | Type | Requirement | Values & limits | Default |
|---|---|---|---|---|
Fieldmodel | Typestring | RequirementRequired | Values & limits qwen/qwen-image-3-pro | Default— |
Fieldprompt | Typestring | RequirementRequired | Values & limits Minimum length: 1 | Default— |
Fieldresolution | Typestring | RequirementOptional | Values & limits 1K, 2K | Default— |
Fieldaspect_ratio | Typestring | RequirementOptional | Values & limits 1:1, 1:2, 1:4, 2:1, 2:3, 3:2, 3:4, 4:1, 4:3, 4:5, 5:4, 9:16, 16:9 | Default— |
Fieldn | Typeinteger | RequirementOptional | Values & limits Minimum: 1 Maximum: 6 | Default— |
Fieldinput_references | Typearray | RequirementOptional | Values & limits Minimum items: 0 Maximum items: 4 | 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 |
| Configurationinput image | Rate schedule— | Forecast quantity— | Billing unitper input image | Charge Input · input image · Estimate for comparison only $0.006 per input image |
| ConfigurationEstimated generation cost | Rate schedule— | Forecast quantityone generation estimate | Billing unit— | Charge Output · Estimated generation cost · Estimate for comparison only $0.08 estimate before generation This estimate is not an additional charge or a fixed rate for the completed generation. |
Ailoot reserves an estimated amount before a request. After the result, it settles the final charge from measured usage. Streaming settlement uses the final usage record. How reservation and settlement work →
set -eu\n: "${AILOOT_API_KEY:?Set your Ailoot API key}"\nexport AILOOT_API_KEY\npython3 - <<'PY_REQUEST'\nimport json,os\nfrom pathlib import Path\np={'model': 'qwen/qwen-image-3-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 qwen/qwen-image-3-pro
/v1/images