Flux Kontext Dev LoRA Ultra Fast API Documentation

Flux Kontext Dev LoRA Ultra Fast API Documentation

Playground

Ultra-fast FLUX.1 Kontext [dev] endpoint with LoRA support for rapid image editing and brand/style adaptation using pre-trained LoRA. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

FLUX Kontext Dev LoRA Ultra Fast is a low-latency image-to-image editing model that supports LoRA adapters directly in the request. Provide a source image plus a natural-language edit instruction, and optionally attach up to 3 LoRAs to steer style, identity consistency, or domain aesthetics—optimized for rapid iteration and production workflows.

Key capabilities

• Ultra-fast instruction-based image editing from a single input image
• LoRA-enabled inference: apply up to 3 LoRAs via input parameters
• Strong preservation when you explicitly state what must remain unchanged
• Great for iterative editing: quick refinements across multiple passes with minimal drift

Typical use cases

• Fast retouching + consistent styling using a “house look” LoRA
• Batch product edits (color variants, background swaps) with brand LoRAs
• Text edits on packaging/signage while keeping typography/perspective consistent
• Rapid A/B testing by switching LoRAs instead of rewriting prompts

Pricing

$0.025 per image.

Cost per run = num_images × $0.025 Example: num_images = 4 → $0.10

Inputs and outputs

Input:

• One source image (upload or public URL)
• One edit instruction (prompt)
• Optional: up to 3 LoRA items

Output:

• One or more edited images (controlled by num_images)

Parameters

Core:

• prompt: Edit instruction describing what to change and what to preserve
• image: Source image
• width / height: Output resolution
• num_inference_steps: More steps can improve fidelity but increases latency
• guidance_scale: Higher values follow the prompt more strongly; too high may over-edit
• num_images: Number of variations generated per run
• seed: Fixed value for reproducibility; -1 for random
• output_format: jpeg or png
• enable_base64_output: Return BASE64 instead of a URL (API only)
• enable_sync_mode: Wait for generation and return results directly (API only)

LoRA (up to 3 items):

• loras: A list of LoRA entries (max 3)

• path: owner/model-name or a direct.safetensors URL

• scale: LoRA strength (start around 0.6–1.0 and adjust)

Prompting guide

Use a clear “preserve + edit + constraints” structure and let LoRAs control the look:

Template: Keep [what must stay]. Change [what to edit]. Ensure [constraints]. Apply LoRA style consistently without altering identity.

Example prompts

• Keep the person’s face, hairstyle, and pose unchanged. Replace the background with a clean studio backdrop. Match lighting direction and shadow softness.
• Keep the product shape and label layout unchanged. Replace only the label text with “WaveSpeedAI”, preserving font style, size, and perspective.
• Remove the background clutter and keep the main subject sharp. Preserve natural skin texture while reducing shine.

Best practices

• Start with one LoRA first; add a second/third only when needed.
• If the output is over-stylized, reduce LoRA scale and/or guidance_scale.
• For consistent batches, reuse the same LoRA set + scales and fix seed for comparisons.
• Match width/height to the original aspect ratio to avoid distortions.

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",
  "num_inference_steps": 28,
  "guidance_scale": 2.5,
  "num_images": 1,
  "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/flux-kontext-dev-lora-ultra-fast" \
  -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