Qwen Image 3.0 Pro

Image GenerationImage Editing1K / 2K Tiers

Qwen Image 3.0 Pro is the higher tier of the Qwen Image 3.0 route on APIXO. It takes the same seven request fields as the base model — text-to-image, image-to-image with up to three reference images, fifteen framing presets, and a 1K or 2K render — but prices the two resolution tiers separately: $0.04 for a 1K image and $0.075 for a 2K image. Draft at 1K, then spend on 2K only for the frames you actually ship.

Modes

Text-to-Image / Image-to-Image

APIXO Price

$0.04 / Image

Resolution

1K / 2K

Reference Images

Up to 3

Output

One Image per Task

Loading workspace...

Qwen Image 3.0 Pro Model API Rate Structure

Billed per successfully generated image, from $0.04 to $0.084. Resolution and the number of reference images both move the rate, so the price of a request is known before you send it.

Text-to-image

1K per image

$0.04

2K per image

$0.075

Image-to-image · 1 reference

1K per image

$0.043

2K per image

$0.078

Image-to-image · 2 references

1K per image

$0.046

2K per image

$0.081

Image-to-image · 3 references

1K per image

$0.049

2K per image

$0.084

Request	1K per image	2K per image
Text-to-image	$0.04	$0.075
Image-to-image · 1 reference	$0.043	$0.078
Image-to-image · 2 references	$0.046	$0.081
Image-to-image · 3 references	$0.049	$0.084

Billing Rules

The 2K tier costs roughly twice the 1K tier on this route. Iterating at 1K and re-rendering the chosen frame at 2K is the cheaper path to the same deliverable.
Every reference image adds $0.003, and image-to-image accepts at most three.
The charge is reserved when a task is accepted and released if the generation does not succeed, so failed jobs do not cost anything.

Common Cost Examples

Example: three 1K drafts plus one 2K final costs $0.195.

Checking API access...

Integrate with AI

Copy a prompt for your AI assistant to install APIXO MCP, read the live schema, and wire this model into your app.

Endpoints

Create task

POSThttps://api.apixo.ai/api/v1/generateTask/qwen-image-3-0-pro

Poll status

GEThttps://api.apixo.ai/api/v1/statusTask/qwen-image-3-0-pro?taskId={taskId}

Request body

request_type

async

Use callback with callback_url for webhooks.

input

object

Mode, prompt, references, size, duration, and audio settings.

request_typerequired

Typeselect

Use async to poll statusTask, or callback to receive a webhook when the result is ready.async · callback

callback_urlconditional

Typetext

Public HTTPS URL for callback delivery.Only in request_type=callback

moderequired

Typeselect

Choose the generation workflow supported by this model.text-to-image · image-to-image

promptrequired

Typestring

Describe the subject, composition, style, lighting, and details you want in the final image.

aspect_ratiorequired

Typeselect

Choose the output frame ratio.auto · 1:1 · 1:2 · 2:1 · 1:3 · 3:1 · 2:3 · 3:2 · 3:4 · 4:3 · 4:5 · 5:4 · 9:16 · 16:9 · 9:21 · 21:9

resolutionrequired

Typeselect

Choose the output resolution tier. Higher tiers can cost more.1k · 2k

enable_prompt_expansionoptional

Typeboolean

Let the model expand a short prompt with additional detail before rendering. Turn it off to keep the wording exactly as written.

seedoptional

Typetext

Use -1 for a random seed, or a whole number from 0 to 2147483647 to rerun the same configuration.

Parameter	Type	Required	Description
`request_type`	select	required	Use async to poll statusTask, or callback to receive a webhook when the result is ready.async · callback
`callback_url`	text	conditional	Public HTTPS URL for callback delivery.Only in request_type=callback
`mode`	select	required	Choose the generation workflow supported by this model.text-to-image · image-to-image
`prompt`	string	required	Describe the subject, composition, style, lighting, and details you want in the final image.
`aspect_ratio`	select	required	Choose the output frame ratio.auto · 1:1 · 1:2 · 2:1 · 1:3 · 3:1 · 2:3 · 3:2 · 3:4 · 4:3 · 4:5 · 5:4 · 9:16 · 16:9 · 9:21 · 21:9
`resolution`	select	required	Choose the output resolution tier. Higher tiers can cost more.1k · 2k
`enable_prompt_expansion`	boolean	optional	Let the model expand a short prompt with additional detail before rendering. Turn it off to keep the wording exactly as written.
`seed`	text	optional	Use -1 for a random seed, or a whole number from 0 to 2147483647 to rerun the same configuration.

File parameters (URL arrays)

image_urls[ mode=image-to-image ]Image-to-image takes one to three reference images. Each image beyond the first adds $0.003 to the price of the generated image.

Only in mode=image-to-image

Example (cURL)

1curl -X POST "https://api.apixo.ai/api/v1/generateTask/qwen-image-3-0-pro" \2  -H "Authorization: Bearer YOUR_API_KEY" \3  -H "Content-Type: application/json" \4  -d '{5    "request_type": "async",6    "input": {7      "mode": "text-to-image",8      "prompt": "A sunlit ceramic studio at golden hour, editorial product photography",9      "aspect_ratio": "4:3",10      "resolution": "2k",11      "enable_prompt_expansion": true,12      "seed": -113    }14  }'

Full API documentation

Capabilities

Seven request fields, two resolution tiers, one image per task

A Separately Priced 2K Tier

Choose 1K at $0.04 or 2K at $0.075 per generated image. The resolution field is a real billing tier here, not a free flag, so the larger render is a deliberate spend rather than something you enable by accident.

Up to Three Reference Images

Image-to-image accepts one to three publicly reachable image URLs alongside the instruction. Each image past the first adds $0.003, so the cost of a multi-reference edit is knowable before you send it.

Framing That Follows the Source

Leave the aspect ratio on Auto in image-to-image and the output follows the proportions of the first reference image, which is the documented behaviour when the field is omitted. Set a ratio explicitly only when you want to depart from the source framing.

Fifteen Framing Presets

Request any of 15 ratios, from square and classic 4:3 through 4:5 and 5:4 social crops to 21:9 and 9:21 ultrawide and 1:3 and 3:1 banner strips.

Optional Prompt Expansion

Prompt expansion is on by default and fills in detail a short brief leaves out. Switch it off when the wording is already exact and you want the model to follow it literally.

Seed Control

Send -1 for a random seed, or any whole number from 0 to 2147483647 to rerun the same configuration while you change one field at a time.

APIXO Workflows

Write a prompt, or hand the model up to three references

Text-to-Image

Send a prompt with an optional framing preset and resolution tier. One image comes back per accepted task: $0.04 at 1K, $0.075 at 2K. Prompt expansion and a seed are the only other knobs.

Create from text

Image-to-Image

Supply one to three reference image URLs plus an instruction of up to 800 characters. Pricing starts at $0.043 at 1K and $0.078 at 2K, and every additional reference image adds $0.003. Leave the ratio on Auto to keep the first reference image’s proportions.

Edit with references

APIXO Configuration

Published request fields

Every control available through APIXO’s Qwen Image 3.0 Pro route.

1–800 Characters

Editing Prompt Length

1 Image

Output per Task

15 Presets

Aspect Ratios

1K / 2K

Resolution Tiers

Up to 3 URLs

Reference Images

Async / Callback

Delivery

Production Uses

Where paying for the 2K tier earns its place

Final Artwork

Render the Frame You Ship at 2K

Take the direction you already settled on and re-render it at the 2K tier for the hero image, cover, or print-bound layout. Because 2K is billed separately at $0.075, you pay for the larger render only on the frames that reach an audience.

Cost Ladder

Explore at 1K, Commit at 2K

Iterate on composition, framing, and wording at $0.04 per image, then switch the resolution field once the direction is locked. The request body is otherwise unchanged, so the promotion step is a single field edit rather than a new integration.

Reference-Led Editing

Combine Up to Three Sources

Give the model a product shot, an environment, and a style plate, then describe the change in 800 characters or fewer. Each additional reference costs $0.003, and leaving the ratio on Auto keeps the first image’s proportions intact.

Layout Systems

Fill Ultrawide and Vertical Slots

Request 21:9 and 9:21 for banners and full-bleed mobile panels, 1:3 and 3:1 for strips, and 4:5 or 5:4 for feed placements — without cropping a square render down to size afterwards.

Notes & FAQ

Billing tiers, prompt limits, and request behavior

Implementation Notes

APIXO bills per successfully generated image. The charge is reserved when the task is accepted and released if the generation does not succeed.

Resolution is a billing tier on this route: $0.04 at 1K and $0.075 at 2K for text-to-image. The base Qwen Image 3.0 route charges the same rate for both.

Image-to-image adds $0.003 for every reference image, up to the documented maximum of three.

Prompts are limited to 800 characters in image-to-image mode. Text-to-image has no documented limit, but the APIXO workspace keeps the same 800-character box for both modes.

Leaving the aspect ratio unset in image-to-image is a documented behavior, not an omission: the output follows the proportions of the first reference image.

Each accepted task returns exactly one image. Submit separate tasks for variations; there is no batch-count field on this route.

Frequently Asked Questions

The request fields are identical; the published price structure is not. Pro prices its two resolution tiers separately — $0.04 at 1K and $0.075 at 2K — while the base route charges one rate regardless of resolution. Choose Pro when the 2K render is the deliverable, and the base route when you are generating in volume and want a flat unit cost.

Published rates run from $0.04 to $0.084 per generated image. Text-to-image is $0.04 at 1K and $0.075 at 2K; image-to-image starts at $0.043 at 1K and $0.078 at 2K, plus $0.003 for each reference image beyond the first.

No. Each accepted task produces exactly one image, and the route exposes no image-count field. Send one task per variation you need; every successful task is billed at its own tier rate.

In image-to-image the output follows the proportions of the first reference image. In text-to-image the model falls back to its documented 1:1 default. The Auto option in the workspace simply omits the field so either behavior applies.

The API takes one to three publicly reachable image URLs. When you drop files into the APIXO workspace they are hosted first and the resulting URLs are sent for you, so both paths reach the same request body.