How many images can I upload, and how large can each file be?

You can upload up to nine JPEG, PNG, or WebP files. Keep each file at or below 10 MB until the limits are aligned.

Does Wan 2.7 Image on Kyeo AI support 4K or thinking mode?

No. The page offers 1K/2K, aspect ratio, watermark, and seed. Technical identifiers: `thinking_mode`.

How does Wan 2.7 Image differ from Pro on Kyeo AI?

The input workflow, four controls, one-image output, and nine-image limit are the same. Use matched inputs to assess results.

How many images does one Wan 2.7 Image request generate?

One. Technical identifiers: `enable_sequential`, `n=1`.

AI image model

Alibaba

Wan

Wan 2.7 Image

Wan 2.7 Image on Kyeo AI handles text-to-image generation and editing with up to nine inputs. Each five-credit request returns one image. The form exposes 1K/2K, text-to-image aspect ratio, watermark, and seed—not image sets, thinking mode, palettes, or box editing.

Capabilities

Text-to-Image, Image-to-Image

Credit cost

5 credits per request

Prompt limit

5,000 characters

Upload limit

9 images

Every request fixes `n=1` and `enable_sequential=false`, so the result count is one.

No-image requests generate an image; one to nine `input_urls` switch the route to instruction-based editing.

The form exposes 1K/2K, eight text-to-image ratios, watermark, and a 0–2147483647 seed.

Image sets, thinking mode, palettes, box editing, and a safety-checker control are unavailable.

Try this model for free View credit pricing Back to model gallery

30-second overview

Wan

5 credits per request

What it does best

This is a reduced integration of the upstream interface's `wan/2-7-image` route for single-image generation and multi-image instruction editing. It describes the fields Kyeo AI sends rather than inheriting the full upstream feature set.

Best for

Testing text-to-image, single-image editing, or multi-image references at a fixed five-credit price, with manual review of composition, text, people, products, and input influence.

At a glance

What is this model like?

Enter a prompt of up to 5,000 characters. With no images, the standard route sends a selected aspect ratio; with one to nine images, it sends `input_urls` and follows their framing. Each request fixes `enable_sequential=false` and `n=1` and costs five credits.

What it does best

Best for

Testing text-to-image, single-image editing, or multi-image references at a fixed five-credit price, with manual review of composition, text, people, products, and input influence.

Why use it on Kyeo AI

Kyeo AI combines generation and editing in one form and shows the fixed price plus the four fields it actually sends, making cost and integration limits visible before submission.

Key facts

Quickly assess whether this model matches your workflow.

Also known as

The same model may appear under different names across documentation and community discussions; this list keeps them easy to search and compare.

Wan2.7 Image

wan/2-7-image

wan2.7-image

Alibaba Wan 2.7 Image

Common search intents

These questions and keywords combine public references with common creative use cases to help you choose faster.

how to use Wan 2.7 Image

Wan 2.7 Image credits per request

Wan 2.7 Image supported file types

Wan 2.7 Image text-to-image ratios

Wan 2.7 Image image editing

Wan 2.7 Image 1K 2K

choose a Wan 2.7 Image aspect ratio

Wan 2.7 Image watermark setting

Wan 2.7 Image maximum input images

Wan 2.7 Image file size limit

choose Wan 2.7 Image or Pro

Selection guide

Use these decision points when choosing a model.

Choose it to test Wan at the lower credit price

The standard workflow costs five credits and Pro costs 12 with the same page controls. Verify one representative prompt or edit first, then decide whether a matched Pro comparison is useful.

Compare another model when 4K must be selectable

The standard Wan route tops out at 2K. Compare Nano Banana Pro when an explicit 4K control is required. A matched Wan Pro test is still possible, but Kyeo AI does not expose 4K there either.

Start with a single-variable check

For generation, hold 2K, 1:1, and one representative prompt constant. For editing, start with one image. Add references or change the ratio only after the result has been verified.

Popular comparisons

Common comparison questions help you find the practical differences quickly.

What is the difference between Wan 2.7 Image and Wan 2.7 Image Pro?

Wan 2.7 Image vs Wan 2.7 Image Pro

What is the difference between Wan 2.7 Image and Qwen Image?

Wan 2.7 Image vs Qwen Image

What is the difference between Wan 2.7 Image and Nano Banana Pro?

Wan 2.7 Image vs Nano Banana Pro

Model comparison

Compare the current model with alternatives at a glance.

Dimension

Wan 2.7 Image

Wan 2.7 Image Pro

Qwen Image

Request workflow

Text-to-image or editing with up to nine inputs; one output

The same input workflow and one-image output

Text-to-image or one image input

Page controls

1K/2K, eight text-to-image ratios, watermark, and seed

The same four controls and ratio options

1K/2K/4K, aspect ratio, and output format

Unavailable controls

Image sets, thinking mode, palettes, and box editing

Image sets and advanced controls are also unavailable

Up to eight inputs, with output format on the page

Credits per request

8 at 1K/2K; 14 at 4K

Review focus

Prompt execution, input influence, text, and fine details

Use matched inputs to assess the seven-credit difference

Prompt execution, text, and output format

Practical usage insights

Practical observations based on real creative workflows.

Standard and Pro expose the same Kyeo AI controls

Both pages fix one output, 1K/2K, and four controls. The current code has no extra switch that proves a Pro quality gain, so any difference should be assessed with matched inputs.

The shared file-size mismatch also applies here

The current workflow reflects this limitation.

Capabilities

One page for generation and editing

Without images, Kyeo AI sends the prompt and `aspect_ratio`. With images, it sends one to nine `input_urls` and omits the ratio. Both modes request one output.

Four controls shape the generation

Resolution is 1K or 2K. Text-to-image ratios are 1:1, 16:9, 4:3, 21:9, 3:4, 9:16, 8:1, and 1:8. The remaining controls toggle watermarking and accept an integer seed.

Wan: Transparent credit use

Kyeo AI does not send `thinking_mode`, `color_palette`, `bbox_list`, or `nsfw_checker`, and it disables `enable_sequential`. Their presence upstream does not make them usable here.

Use cases

Text-to-image parameter check

Use a fixed prompt, 2K setting, ratio, and seed to review composition, text, and details. A seed helps control the test.

Single-image instruction editing

Upload one source and state what must stay and what should change. Review the result against the source for object, text, face, and layout errors.

Multi-image reference editing

Provide a small set of subject, material, or composition references and identify each role.

Prompt tips

Fix the ratio and resolution first

Change only the prompt in the first pass. Simultaneous changes to aspect ratio, resolution, and seed make the result difficult to attribute.

Assign a role to each reference image

State which image supplies the subject, material, composition, or background, and add references one at a time. Filling all nine slots makes conflicts harder to manage.

Set delivery constraints before generating

The page offers 1K/2K and four controls. Base customer commitments, quotes, and advertising on the selected settings and actual result, with human quality and rights review.

Strengths

One page supports text-to-image and editing with up to nine inputs.

Eight ratios cover common landscape, portrait, ultra-wide, and ultra-tall canvases.

The one-image result and five-credit cost are visible before submission.

Matching controls with Pro make controlled comparisons straightforward.

Limitations

Only 1K/2K is available; direct-API 4K is not exposed in this form.

Output is fixed to one image, with no one-to-four output or 12-image set mode.

Thinking mode, palettes, box editing, output count, and safety-checker controls are unavailable.

The documented 10 MB per-image limit is lower than Kyeo's current 20 MB uploader limit.

Adjustable parameters

Quickly assess whether this model matches your workflow.

Aspect ratio

aspect_ratio

Optional

Parameter type: select

Default: 1:1

1:1

16:9

4:3

21:9

3:4

9:16

8:1

1:8

8 options

Resolution

resolution

Optional

Parameter type: select

Default: 2K

Watermark

watermark

Optional

Parameter type: select

Default: false

Off

Random seed

seed

Optional

Parameter type: number

Range 0 - 2147483647

Credit usage

5 credits per request

Kyeo AI charges five credits per request for both text-to-image generation and image editing.

Budget tip

Before batch generation, run an A/B test with the same assets across the current model and alternatives to validate quality and avoid wasting credits.

FAQ

Related models

Compare these similar candidates before deciding.

Wan 2.7 Image Pro

Wan 2.7 Image Pro on Kyeo AI handles text-to-image generation and editing with up to nine inputs. Each 12-credit request returns one image. The form exposes 1K/2K, text-to-image aspect ratio, watermark, and seed—not direct-API 4K, image sets, thinking mode, palettes, or box editing.

AI image model

12 credits per request

View details Use now

Qwen Image

Qwen Image provides prompt-only generation and one-image editing on Kyeo for a fixed 4 credits. Generation exposes six `image_size` presets. Uploading one image switches to the editing route and hides size; both modes offer PNG or JPEG. Prompts may contain up to 5,000 characters, and an uploaded image may be no larger than 10 MB.

AI image model

4 credits per request

View details Use now

Nano Banana Pro

Nano Banana Pro supports text-to-image generation and reference-image editing on Kyeo AI, with 1K, 2K, or 4K output. Each 1K or 2K request costs 8 credits, while 4K costs 14. Resolution is an output-size tier, not a guarantee of quality, accuracy, or detail.

AI image generation and editing model

8–14 credits per request

View details Use now

Sources

Content is based on model documentation, feature references, and the settings available on Kyeo AI.

Source notes

Kyeo AI fixes `n=1` and `enable_sequential=false`, exposes only `aspect_ratio`, `resolution`, `watermark`, and `seed`, and sends up to nine `input_urls` when images are present. It does not expose image sets, thinking mode, color palettes, or bounding-box editing. That shared mismatch is already recorded under the current behavior.

Last reviewed: 2026-07-20

Alibaba Cloud Model Studio: Wan 2.7 image generation and editing API reference

Official

View source

Alibaba Cloud Model Studio: Image generation and editing

Official

View source

Upstream interface documentation: Wan 2.7 Image

Platform

Its example includes `input_urls`, output count, image-set mode, resolution, thinking mode, watermark, seed, and box fields.

upstream interface Wan 2.7 Image product form

Platform

The current form lists a 5,000-character prompt, up to nine 10 MB images, eight ratios, and 1K/2K.

Kyeo AI Wan 2.7 Image configuration and request path

Platform

Kyeo AI uses `wan/2-7-image`, fixes `n=1`, disables image sets, offers 1K/2K, eight ratios, watermark, and seed, accepts nine images, and charges five credits.