Browse ModelsClarity AIClarity AI Creative Upscaler

Clarity Ai Creative Upscaler

Clarity Ai Creative Upscaler

Playground

Try it on WavespeedAI!

Creative Upscaler enlarges images with style and detail control, restoring photos or adding micro-textures for portraits and anime. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

Clarity AI Creative Upscaler

Clarity AI Creative Upscaler enlarges and enhances images while giving you strong control over style, fidelity, and invented detail. It can produce faithful restorations or more stylized, generative upscales for portraits, illustrations, anime, and other creative image workflows.


Why Choose This?

  • Creative super-resolution Upscale images while improving sharpness, cleaning artifacts, and adding richer visual detail.

  • Style-aware enhancement Choose from default, portrait, or anime modes depending on the content type and desired look.

  • Fine-grained control Tune creativity, resemblance, dynamic, and fractality to balance faithfulness, stylization, contrast, and detail density.

  • Prompt-guided refinement Optionally use a short prompt to steer mood, color, texture, or visual tone.

  • Predictable output tiers Choose from 2k, 4k, or 8k target resolution presets for consistent output sizing.

  • Production-ready API Suitable for enhancement pipelines, creative tooling, artwork preparation, portrait workflows, and commercial asset production.


Parameters

ParameterRequiredDescription
imageYesSource image URL or uploaded image. Better source quality generally produces better upscales.
target_resolutionNoOutput resolution tier: 2k, 4k (default), or 8k. Higher tiers create larger and more detailed images.
promptNoOptional prompt to guide mood, texture, tone, color, or enhancement direction.
styleNoStyle preset: default, portrait, or anime. Default: default.
creativityNoControls how much new detail the model invents. Range: -10 to 10. Default: 0.
resemblanceNoControls how strongly the output preserves the original content, identity, and composition. Range: -10 to 10. Default: 0.
dynamicNoControls overall contrast and visual pop. Range: -10 to 10. Default: 0.
fractalityNoControls fine detail density and texture complexity. Range: -10 to 10. Default: 0.

How to Use

  1. Upload your image — provide the source image you want to upscale.
  2. Choose target resolution — select 2k, 4k, or 8k depending on your quality and cost needs.
  3. Choose a style preset — use default, portrait, or anime based on the image type.
  4. Add a prompt (optional) — use a short directive if you want to guide texture, tone, lighting, or mood.
  5. Adjust enhancement controls (optional) — tune creativity, resemblance, dynamic, and fractality based on how faithful or stylized you want the output to be.
  6. Submit — run the model and download the enhanced image.

Example Prompt

Warm film look, soft skin texture, subtle grain, golden hour lighting, natural contrast


Pricing

Pricing is based on the selected target_resolution.

Target ResolutionCost
2k$0.05
4k$0.10
8k$0.20

Billing Rules

  • 2k costs $0.05 per image
  • 4k costs $0.10 per image
  • 8k costs $0.20 per image
  • Pricing depends on target_resolution
  • prompt, style, creativity, resemblance, dynamic, and fractality do not affect pricing

Best Use Cases

  • Portrait enhancement — Upscale faces with more controlled skin handling and gentler restoration using portrait style.
  • Anime and illustration upscaling — Preserve line integrity and flat-color structure while increasing resolution with anime style.
  • Creative photo enhancement — Add richer textures, stronger mood, and stylized detail to photographs.
  • Artwork preparation — Generate larger, cleaner assets for digital art, posters, prints, and portfolios.
  • Commercial asset refinement — Improve image quality for banners, campaigns, presentations, and product visuals.
  • Controlled restoration vs stylization — Fine-tune whether the result should stay close to the original or become more interpretive.

Pro Tips

  • Start with 4k for most general use cases.
  • Keep creativity lower when preserving faces, logos, or important shapes matters.
  • Raise resemblance when identity, composition, and structure need to stay closer to the source.
  • Use portrait style for people and anime style for line art, cel shading, or stylized illustrations.
  • Increase fractality carefully, especially on natural photos, since too much can make textures look artificial.
  • Reduce dynamic if highlights become too punchy or shadows feel too heavy.
  • Leave prompt blank for more neutral results, or use a short and specific prompt for better control.

Notes

  • image is the only required field.
  • Supported target resolutions are 2k, 4k, and 8k.
  • Supported styles are default, portrait, and anime.
  • target_resolution defaults to 4k.
  • creativity, resemblance, dynamic, and fractality each range from -10 to 10.
  • Pricing depends only on the selected resolution tier.

Authentication

For authentication details, please refer to the Authentication Guide.

API Endpoints

Submit Task & Query Result


# Submit the task
curl --location --request POST "https://api.wavespeed.ai/api/v3/clarity-ai/creative-upscaler" \
--header "Content-Type: application/json" \
--header "Authorization: Bearer ${WAVESPEED_API_KEY}" \
--data-raw '{
    "target_resolution": "4k",
    "style": "default",
    "creativity": 0,
    "resemblance": 0,
    "dynamic": 0,
    "fractality": 0
}'

# Get the result
curl --location --request GET "https://api.wavespeed.ai/api/v3/predictions/${requestId}/result" \
--header "Authorization: Bearer ${WAVESPEED_API_KEY}"

Parameters

Task Submission Parameters

Request Parameters

ParameterTypeRequiredDefaultRangeDescription
imagestringYes-URL to the input image,mast be url encoded
target_resolutionstringNo4k2k, 4k, 8kTarget output resolution tier. Higher tiers create larger, more detailed images.
promptstringNo-The prompt for the generation
stylestringNodefaultdefault, portrait, animestyle:default, portrait, or anime
creativitynumberNo--10 ~ 10value between -10 and 10
resemblancenumberNo--10 ~ 10value between -10 and 10
dynamicnumberNo--10 ~ 10value between -10 and 10
fractalitynumberNo--10 ~ 10value between -10 and 10

Response Parameters

ParameterTypeDescription
codeintegerHTTP status code (e.g., 200 for success)
messagestringStatus message (e.g., “success”)
data.idstringUnique identifier for the prediction, Task Id
data.modelstringModel ID used for the prediction
data.outputsarrayArray of URLs to the generated content (empty when status is not completed)
data.urlsobjectObject containing related API endpoints
data.urls.getstringURL to retrieve the prediction result
data.statusstringStatus of the task: created, processing, completed, or failed
data.created_atstringISO timestamp of when the request was created (e.g., “2023-04-01T12:34:56.789Z”)
data.errorstringError message (empty if no error occurred)
data.timingsobjectObject containing timing details
data.timings.inferenceintegerInference time in milliseconds

Result Request Parameters

ParameterTypeRequiredDefaultDescription
idstringYes-Task ID

Result Response Parameters

ParameterTypeDescription
codeintegerHTTP status code (e.g., 200 for success)
messagestringStatus message (e.g., “success”)
dataobjectThe prediction data object containing all details
data.idstringUnique identifier for the prediction, the ID of the prediction to get
data.modelstringModel ID used for the prediction
data.outputsstringArray of URLs to the generated content (empty when status is not completed).
data.urlsobjectObject containing related API endpoints
data.urls.getstringURL to retrieve the prediction result
data.statusstringStatus of the task: created, processing, completed, or failed
data.created_atstringISO timestamp of when the request was created (e.g., “2023-04-01T12:34:56.789Z”)
data.errorstringError message (empty if no error occurred)
data.timingsobjectObject containing timing details
data.timings.inferenceintegerInference time in milliseconds
© 2025 WaveSpeedAI. All rights reserved.