Clarity AI Flux Upscaler

Clarity AI Flux Upscaler

Playground

Try it on WaveSpeedAI!

Clarity AI Flux Upscaler sharpens images while preserving natural textures and edges, with prompt-guided refinement and LoRA support. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

Clarity AI Flux Upscaler enhances and enlarges images with prompt-guided refinement, optional LoRA style control, and adjustable creativity. It is designed for high-quality super-resolution workflows where you want both sharper detail and more control over the final look.


Why Choose This?

  • Prompt-guided upscaling Use a text prompt to steer texture, tone, lighting, or style during enhancement.

  • Optional LoRA control Add a compatible lora_link when you want stronger domain-specific styling.

  • Megapixel-based output sizing Choose the target output size directly in megapixels for predictable delivery.

  • Flexible enhancement strength Adjust creativity to balance faithful restoration against more generative detail.

  • Production-ready workflow Suitable for portraits, fashion, product imagery, artwork, and other high-resolution creative assets.


Parameters

ParameterRequiredDescription
imageYesInput image to upscale.
target_megapixelsNoTarget output size in megapixels. Higher values produce larger and more detailed outputs.
promptNoOptional prompt to guide tone, texture, lighting, or visual refinement.
lora_linkNoOptional URL to a compatible LoRA for additional style control.
creativityNoControls how much new detail is added. Lower values stay closer to the source, while higher values add stronger enhancement.

How to Use

  1. Upload your image — provide the source image you want to enhance.
  2. Choose target megapixels — set the desired output size based on your delivery needs.
  3. Add a prompt (optional) — describe the look, texture, or mood you want.
  4. Add a LoRA link (optional) — provide lora_link if you want extra style steering.
  5. Adjust creativity (optional) — keep it low for more faithful results, or raise it for stronger generated detail.
  6. Submit — run the model and download the enhanced image.

Example Prompt

Elegant editorial texture, soft natural light, refined fabric detail, realistic skin tones, premium fashion photography look


Pricing

Pricing is based on the selected target_megapixels tier.

Target MegapixelsCost
<= 4 MP$0.20
> 4 MP and <= 8 MP$0.40
> 8 MP and <= 16 MP$0.60
> 16 MP and <= 25 MP$1.20
> 25 MP and <= 50 MP$2.40
> 50 MP$3.20

Billing Rules

  • Pricing is based on target_megapixels
  • Cost uses fixed megapixel tiers rather than scaling linearly
  • Moving into a higher megapixel tier increases the price to the next bracket
  • prompt, lora_link, and creativity do not affect pricing

Best Use Cases

  • Fashion and portrait enhancement — Improve detail while preserving skin tones, fabrics, and visual polish.
  • Product image refinement — Generate sharper commercial assets for catalogs, campaigns, and listings.
  • Artwork and illustration upscaling — Create larger outputs with more controlled stylization.
  • Prompt-guided image finishing — Use short prompts to nudge mood, texture, and overall visual tone.
  • Style-directed super-resolution — Combine megapixel scaling with LoRA guidance for custom looks.
  • Premium asset preparation — Produce high-resolution images for design, publishing, and creative delivery.

Pro Tips

  • Start with a lower megapixel tier first, then increase only if you need a larger final output.
  • Keep creativity lower when identity, structure, and source fidelity matter most.
  • Use short, specific prompts instead of long descriptive paragraphs.
  • Add lora_link only when you need a stronger style direction.
  • Use the cleanest source image available for better detail recovery.
  • For fashion, portrait, or editorial work, subtle prompts usually produce more controllable results.

Notes

  • image is the only required field.
  • Pricing depends on the selected target_megapixels tier.
  • prompt, lora_link, and creativity change the look of the result, but not the price.
  • Larger target sizes are better suited for print, premium delivery, and high-resolution commercial use.

Authentication

For authentication details, please refer to the Authentication Guide.

API Endpoints

Submit Task & Query Result

set -euo pipefail

export WAVESPEED_API_KEY="your-api-key"

REQUEST_BODY=$(cat <<'JSON'
{
  "image": "https://interactive-examples.mdn.mozilla.net/media/cc0-images/painted-hand-298-332.jpg",
  "target_megapixels": 4,
  "creativity": 0
}
JSON
)

# 1. Submit the prediction.
SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  -X POST "https://api.wavespeed.ai/api/v3/clarity-ai/flux-upscaler" \
  -H "Authorization: Bearer ${WAVESPEED_API_KEY}" \
  -H "Content-Type: application/json" \
  -d "${REQUEST_BODY}")

TASK=$(printf '%s' "${SUBMIT_RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
PREDICTION_ID=$(printf '%s' "${TASK}" | jq -r '.id // empty')
if [ -z "${PREDICTION_ID}" ]; then
  printf 'Submission response did not contain a prediction id
' >&2
  exit 1
fi
RESULT_URL=$(printf '%s' "${TASK}" | jq -r '.urls.get // empty')
if [ -z "${RESULT_URL}" ]; then RESULT_URL="https://api.wavespeed.ai/api/v3/predictions/${PREDICTION_ID}/result"; fi

# 2. Poll until the prediction finishes.
while true; do
  RESPONSE=$(curl --silent --show-error --fail-with-body \
    "${RESULT_URL}" \
    -H "Authorization: Bearer ${WAVESPEED_API_KEY}")
  RESULT=$(printf '%s' "${RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
  STATUS=$(printf '%s' "${RESULT}" | jq -r '.status // empty')

  case "${STATUS}" in
    completed) printf '%s\n' "${RESULT}" | jq '.outputs'; break ;;
    failed|cancelled|timeout) printf '%s\n' "${RESULT}" | jq . >&2; exit 1 ;;
    created|processing) sleep 2 ;;
    *) printf 'Unexpected status: %s
' "${STATUS}" >&2; exit 1 ;;
  esac
done

Parameters

Task Submission Parameters

Request Parameters

ParameterTypeRequiredDefaultRangeDescription
imagestringYes-Input image to upscale.
target_megapixelsnumberNo41 ~ 64Requested output size in megapixels. Range: 1-64 MP.
creativitynumberNo0-10 ~ 10Negative values stay stricter to the source; positive values add more generated detail.

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.outputsarrayOutput values, usually URL strings; some models return text strings or structured result objects (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
data.modelstringModel ID used for the prediction
data.outputsarray<string | object>Array of generated outputs (empty when status is not completed). Items are usually URL strings, but may be text strings or structured result objects, depending on the model.
data.urlsobjectObject containing related API endpoints
data.urls.getstringURL to poll for the prediction result
data.statusstringStatus: created, processing, completed, or failed
data.created_atstringISO timestamp of when the request was created
data.errorstringError message (empty if no error occurred)
data.timingsobjectObject containing timing details
data.timings.inferenceintegerInference time in milliseconds
© 2026 WaveSpeedAI. All rights reserved.