Moss Image Text To Image LoRA API Documentation

Moss Image Text To Image LoRA API Documentation

Playground

MOSS Image Text-to-Image LoRA generates images from long text prompts with up to 3 custom LoRAs, with sharp Chinese and English text rendering, 1K, 1.5K or 2K output and flexible aspect ratios. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

MOSS Image Text-to-Image LoRA generates images from detailed text prompts with strong Chinese and English text rendering, while supporting up to 3 custom LoRA weights in a single request. It is well suited for posters, packaging, infographics, and other prompt-driven visuals that need both visual control and readable text.

Why Choose This?

• Detailed prompt understanding — Use rich prompts to control subject, composition, environment, lighting, and style.
• Chinese and English text rendering — Generate signs, posters, packaging, and other text-heavy images with clearer readable text.
• Flexible resolution options — Choose 1k, 1.5k, or 2k depending on your quality and cost needs.
• LoRA support — Apply up to 3 LoRAs in one request for stronger style or subject control.
• Flexible framing — Combine different aspect ratios with multiple output resolutions.
• Multiple output formats — Export as jpeg, png, or webp.

Parameters

How to Use

1. Write your prompt — Describe the subject, layout, background, lighting, and style. Put any text you want rendered in quotation marks.
2. Choose a resolution — Select 1k, 1.5k, or 2k.
3. Add LoRAs if needed — Use up to 3 LoRAs, each with its own scale.
4. Choose framing and format — Optionally set aspect_ratio and output_format.
5. Set a seed if needed — Use a fixed seed for more reproducible results, or -1 for random generation.
6. Submit — Generate and retrieve the final image.

Pricing

The LoRA surcharge is already included in the prices above. Aspect ratio, output format, and seed do not add separate charges.

Best Use Cases

• Posters and promotional graphics
• Packaging and label concepts
• Infographics with Chinese or English text
• Advertising and marketing creatives
• Editorial and social media visuals
• Style-controlled image generation with LoRAs

Pro Tips

• Be specific about the subject, composition, lighting, and style.
• Put any text that should appear in the image inside quotation marks.
• Use LoRAs when you want stronger control over style, character, or visual identity.
• Start with 1k for quick testing, then move to higher resolutions for final output.
• Use a fixed seed when comparing prompt or LoRA changes.

Related Models

• MOSS Image Text-to-Image
• MOSS Image Text-to-Image LoRA

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'
{
  "prompt": "A cinematic ocean wave at sunrise, highly detailed",
  "aspect_ratio": "1:1",
  "resolution": "1k",
  "output_format": "jpeg"
}
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/wavespeed-ai/moss-image/text-to-image-lora" \
  -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

Every model also accepts the global request options enable_sync_mode and enable_base64_output; see their pages for details.

Response Parameters

Result Request Parameters

Result Response Parameters