Bria Virtual Try On API Documentation

Bria Virtual Try On API Documentation

Playground

Bria Virtual Try-On dresses a person in reference garments using a person photo, one to three garment or accessory reference images, and optional text instructions for virtual try-on, fashion visualization, e-commerce, and styling workflows. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

Bria Virtual Try-On lets you visualize garments and accessories on a person using a portrait or full-body photo and one to three reference images. Upload the person image separately from the clothing or accessory references, then optionally add instructions to guide how the items should appear in the final result.

The model is designed for fashion visualization, apparel previews, accessory try-on, and e-commerce workflows where you want to place referenced items onto an existing person image.

Why Choose This?

• Purpose-built editing
  Visualize garments and accessories on a person from reference images.

• Multi-reference support
  Combine one person photo with one to three garment or accessory reference images.

• Optional prompt guidance
  Images alone are sufficient. Add a prompt only when you want to direct a specific edit.

• Original framing by default
  Omit aspect_ratio to retain the person photo’s aspect ratio.

• Standard image output
  Receive a PNG or JPEG image URL in the standard WaveSpeed prediction response.

Parameters

How to Use

1. Upload a person photo - Provide the base person image in image.
2. Add reference images - Supply one to three garment or accessory images in reference_images.
3. Guide the edit optionally - Add a concise prompt for placement or styling details.
4. Choose output settings - Optionally set the aspect ratio, output format, and seed.
5. Submit - Poll the prediction result and retrieve the generated image URL.

Pricing

Pricing is fixed at $0.04 per generated image. Each successful request returns one image.

One to three reference images are included at the same price. prompt, aspect_ratio, output_format, and seed do not add separate charges.

Best Use Cases

• Virtual clothing try-on
• Fashion catalog variations
• Multi-item outfit visualization

Pro Tips

• Use a clear person photo with the relevant body area visible.
• Use clear garment references with visible fabric, patterns, and closures. Use the prompt to explain layering or which existing items to retain.
• Leave aspect_ratio unset when you want to preserve the original framing.
• Keep all input image URLs publicly accessible while the request is processing.
• Use a fixed seed when comparing prompt changes.

Notes

• image and reference_images are required; prompt is optional.
• The limit is one person photo plus one to three reference images, for two to four images in total.

Related Models

• Bria Virtual Try-On — Apply garment or accessory references to an existing person photo for virtual try-on workflows.
• Bria Product Holding — Generate images of a person naturally holding or presenting a referenced product.

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",
  "reference_images": [
    "https://interactive-examples.mdn.mozilla.net/media/cc0-images/painted-hand-298-332.jpg"
  ],
  "aspect_ratio": "1:1",
  "output_format": "png"
}
JSON
)

# 1. Submit the prediction once. Never retry this request automatically:
# a retry after an unclear failure can create a second, billed task.
CHECK='The task may already exist. Check History or the predictions list before submitting again.'
if ! SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  --connect-timeout 10 --max-time 60 \
  -X POST "https://api.wavespeed.ai/api/v3/bria/virtual-try-on" \
  -H "Authorization: Bearer ${WAVESPEED_API_KEY}" \
  -H "Content-Type: application/json" \
  -d "${REQUEST_BODY}"); then
  printf 'Submit failed. %s\n' "${CHECK}" >&2
  exit 1
fi

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 "${PREDICTION_ID}" ]; then
  printf 'Submission response did not contain a prediction id. %s\n' "${CHECK}" >&2
  exit 1
fi
RESULT_URL="https://api.wavespeed.ai/api/v3/predictions/${PREDICTION_ID}/result"

# 2. Poll until the prediction reaches a final status.
BODY_FILE=$(mktemp)
trap 'rm -f "${BODY_FILE}"' EXIT
while true; do
  # Transient errors (429, 5xx) are retried, honoring Retry-After.
  HTTP_CODE=$(curl --silent --show-error --fail-with-body \
    --connect-timeout 10 --max-time 60 --retry 5 --retry-delay 2 \
    --output "${BODY_FILE}" --write-out '%{http_code}' \
    "${RESULT_URL}" \
    -H "Authorization: Bearer ${WAVESPEED_API_KEY}" || true)
  case "${HTTP_CODE}" in
    2??) ;;
    000|429|5??)  # dropped connection or temporary error: keep polling
      printf 'Task %s: HTTP %s, polling again in 10s\n' "${PREDICTION_ID}" "${HTTP_CODE}" >&2
      sleep 10; continue ;;
    *) printf 'Task %s: HTTP %s\n' "${PREDICTION_ID}" "${HTTP_CODE}" >&2
      cat "${BODY_FILE}" >&2; exit 1 ;;
  esac
  RESULT=$(jq 'if type == "object" and has("data") then .data else . end' "${BODY_FILE}")
  STATUS=$(printf '%s' "${RESULT}" | jq -r '.status // empty')

  case "${STATUS}" in
    completed) printf '%s\n' "${RESULT}" | jq '.outputs'; break ;;
    failed|cancelled|timeout|deleted)
      printf 'Task %s ended with status %s\n' "${PREDICTION_ID}" "${STATUS}" >&2
      printf '%s\n' "${RESULT}" | jq . >&2; exit 1 ;;
    *) sleep 2 ;;
  esac
done

Parameters

Task Submission Parameters

Request Parameters

Response Parameters

Result Request Parameters

Result Response Parameters