Pruna AI P Image Text To Image LoRA API Documentation

Pruna AI P Image Text To Image LoRA API Documentation

Playground

Pruna AI P-Image Text to Image LORA is a fast AI image generation model that creates high-quality images from text prompts with LORA-based customization. Ready-to-use REST inference API for stylized image generation, brand-specific visuals, character design, product concepts, marketing creatives, and custom AI image workflows with simple integration, no coldstarts, and affordable pricing.

Features

Pruna AI P-Image Text-to-Image LoRA generates images from natural-language prompts while allowing you to apply a custom LoRA for style or subject control. It is designed for workflows where you want the flexibility of text-to-image generation together with a Pruna-trained LoRA for more specialized visual outputs.

Why Choose This?

• LoRA-powered image generation Generate images from prompts while steering the result with a custom LoRA.

• Custom style and subject control Use lora_weights to apply a specialized visual style, character concept, or trained aesthetic.

• Flexible sizing options Choose a preset aspect_ratio or switch to custom for direct width and height control.

• LoRA strength adjustment Use lora_scale to control how strongly the LoRA influences the final image.

• Seed support for reproducibility Reuse the same seed to generate more consistent variations.

• Affordable fixed pricing Each generation run uses a simple flat per-image price.

Parameters

How to Use

1. Write your prompt — describe the subject, composition, style, lighting, and mood you want.
2. Provide your LoRA — add lora_weights pointing to the LoRA you want to use.
3. Adjust LoRA strength (optional) — set lora_scale to control how strongly the LoRA affects the result.
4. Choose aspect ratio — use a preset ratio or select custom if you want direct control over width and height.
5. Set width and height (optional) — these only take effect when aspect_ratio is set to custom.
6. Add a Hugging Face token (optional) — use hf_api_token if your LoRA is hosted in a private Hugging Face repository.
7. Set a seed (optional) — use a fixed seed for more reproducible generations.
8. Submit — run the model and download the generated image.

Example Prompt

comic noir art style, a detective standing under a street lamp in the rain

Pricing

Just $0.005 per image.

Best Use Cases

• Custom style generation — Apply a trained LoRA to generate images in a specific visual style.
• Character and concept generation — Use LoRAs trained for specific characters, identities, or art directions.
• Brand or aesthetic consistency — Keep outputs aligned with a known visual language.
• Creative prototyping — Explore prompt variations with a reusable LoRA model.
• Specialized artwork generation — Generate images for niche styles that standard prompting alone may not capture as reliably.

Pro Tips

• Make sure your lora_weights come from a LoRA trained with the Pruna T2I Trainer.
• Keep prompts clear and focused so the LoRA can guide the result more effectively.
• Adjust lora_scale gradually to find the right balance between prompt influence and LoRA influence.
• Use custom aspect ratio only when you need exact width and height control.
• If your LoRA is stored in a private Hugging Face repo, provide hf_api_token so it can be accessed properly.
• Reuse the same seed when you want more consistent iterations of the same concept.

Notes

• prompt is required.
• width and height only take effect when aspect_ratio is set to custom.
• lora_weights must point to a LoRA trained by the Pruna T2I Trainer.
• hf_api_token is only needed when the LoRA repository is private or otherwise requires authentication.
• Pricing is fixed at $0.005 per image.

Tested Example

{

“prompt”: “comic noir art style, a detective standing under a street lamp in the rain”,

“lora_weights”: “huggingface.co/PrunaAI/p-image-comic-noir-art-lora/weights.safetensors”,

“lora_scale”: 0.8,

“aspect_ratio”: “1:1”,

“output_format”: “png” }

Related Models

• Pruna AI P-Image Text-to-Image — Generate images directly from prompts without a LoRA.
• Pruna AI P-Image Edit — Edit existing images using natural-language instructions.
• Pruna AI P-Image Upscale — Enhance and enlarge existing images.

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",
  "width": 512,
  "height": 512,
  "lora_scale": 0.5,
  "aspect_ratio": "custom",
  "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/pruna-ai/p-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