MiniMax H3 Open Weights | Try in Video Generator →

Flux 2 Klein Base 9B Edit LoRA | Custom LoRA Image API

wavespeed-ai/

FLUX.2 [klein] Base 9B Edit with LoRA support is a high-quality image editing model with 9B parameters, offering precise modifications using natural language instructions and personalized styles via custom LoRA adapters. Ready-to-use REST inference API, best performance, no cold starts, affordable pricing.

lora-support
Input

Idle

Change the atmosphere into the moment before a storm. Preserve the same man, face, pose, outfit, beach layout, camera angle, and composition. Add dark dramatic clouds, stronger wind, rougher ocean waves, and cooler cinematic lighting. Keep the subject realistic and stable.

$0.026per run·~38 / $1

ExamplesView all

Change the atmosphere into the moment before a storm. Preserve the same man, face, pose, outfit, beach layout, camera angle, and composition. Add dark dramatic clouds, stronger wind, rougher ocean waves, and cooler cinematic lighting. Keep the subject realistic and stable.

Change the atmosphere into the moment before a storm. Preserve the same man, face, pose, outfit, beach layout, camera angle, and composition. Add dark dramatic clouds, stronger wind, rougher ocean waves, and cooler cinematic lighting. Keep the subject realistic and stable.

Related Models

README

WaveSpeed AI FLUX.2 Klein Base 9B Edit LoRA

WaveSpeed AI FLUX.2 Klein Base 9B Edit LoRA is a high-quality image editing model with full LoRA support. Built on a 9B-parameter architecture, it delivers stronger detail, better prompt understanding, and more refined edits than the 4B variant. Upload one or more source images, describe the edit in natural language, and optionally apply custom LoRA adapters for personalized styles or character consistency.

Why Choose This?

  • Higher-quality editing The 9B model produces richer detail, stronger prompt adherence, and better overall edit quality than the 4B variant.

  • Natural-language editing Describe the change you want in plain language — transform style, modify content, add effects, or refine the scene.

  • Full LoRA support Apply custom LoRA adapters for personalized styles, characters, aesthetics, or branded visual directions.

  • Multi-image support Upload multiple source images for more context-aware editing and compositing workflows.

  • Flexible output sizing Optionally set output dimensions, or leave size empty to preserve the original input dimensions.

  • Prompt Enhancer Built-in prompt enhancement can help improve edit quality and prompt clarity.

  • Production-ready workflow Suitable for high-quality retouching, style transfer, compositing, and more advanced creative editing tasks.

Parameters

ParameterRequiredDescription
promptYesText description of the desired edit.
imagesYesSource images to edit. Multiple images are supported.
lorasNoList of LoRA adapters to apply during editing.
sizeNoOutput dimensions. Leave empty to match the input image dimensions.
seedNoRandom seed for reproducibility. Use -1 for random generation.

LoRA Format

Each item in the loras array supports:

FieldRequiredDescription
pathYesURL to the LoRA weights file.
scaleNoLoRA weight multiplier. Default: 1.

How to Use

  1. Write your prompt — describe the edit you want, such as “make it a real picture” or “add sunset lighting.”
  2. Upload your images — add one or more source images using the image input.
  3. Add LoRAs (optional) — include one or more LoRA adapters if you want custom style or character control.
  4. Adjust LoRA scale (optional) — start with 1 and fine-tune if needed.
  5. Set size (optional) — specify output dimensions, or leave it empty to preserve the original dimensions.
  6. Set seed (optional) — use -1 for random generation, or enter a fixed seed for reproducible results.
  7. Submit — run the model and download the edited image.

Example Prompt

Turn this illustration into a realistic cinematic portrait, keep the same composition and facial features, add soft sunset lighting and natural skin texture.

Pricing

ItemCost
Per image$0.026

Billing Rules

  • Pricing is fixed at $0.026 per generated image
  • size, seed, and the number of LoRAs do not affect pricing
  • Flat-rate pricing applies regardless of image dimensions or LoRA count

Best Use Cases

  • Style transfer — Transform images using custom LoRA styles or aesthetics.
  • Reality enhancement — Convert illustrations, renders, or stylized images into more photorealistic results.
  • Character consistency — Apply character LoRAs while editing to maintain a recognizable identity.
  • High-quality retouching — Use the 9B model when the 4B variant is not sufficient for professional work.
  • Production editing — Balance strong detail and prompt fidelity with practical cost and speed.
  • Complex image refinement — Handle edits that need stronger scene understanding and more precise visual control.

Pro Tips

  • Be specific about what should change and what should stay the same.
  • Use LoRAs when you want a consistent style or character look across multiple edits.
  • Leave size empty when you want to preserve the original image dimensions.
  • Start with LoRA scale = 1 and adjust based on how strongly you want the adapter to influence the result.
  • Use the same seed when comparing different prompts or LoRA combinations.
  • For best results, use high-quality source images and clearly written prompts.

Notes

  • Both prompt and images are required.
  • If size is not specified, the output matches the input image dimensions.
  • LoRAs can be stacked for combined effects.
  • The 9B model offers better detail and prompt understanding than the 4B variant at a slightly higher cost.
  • For best results, use high-quality source images.

Related Models

Note:This website uses AI models provided by third parties.

Flux 2 Klein Base 9b Edit Lora API — Quick start

Grab a WaveSpeedAI API key, then call POST https://api.wavespeed.ai/api/v3/wavespeed-ai/flux-2-klein-base-9b/edit-lora with your input as JSON. The endpoint returns a prediction id. Start polling the result endpoint around every 2 seconds, increase the interval for long-running tasks, and stop on any terminal status. On completed, read output values from data.outputs. Examples for Flux 2 Klein Base 9b Edit Lora below.

HTTP example
set -euo pipefail

: "${WAVESPEED_API_KEY:?Set WAVESPEED_API_KEY}"

REQUEST_BODY=$(cat <<'JSON'
{
    "prompt": "A cinematic shot of a city at sunset, soft golden light",
    "images": [
        "https://interactive-examples.mdn.mozilla.net/media/cc0-images/painted-hand-298-332.jpg"
    ],
    "seed": -1
}
JSON
)

# 1. Submit the prediction.
SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  -X POST "https://api.wavespeed.ai/api/v3/wavespeed-ai/flux-2-klein-base-9b/edit-lora" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $WAVESPEED_API_KEY" \
  -d "$REQUEST_BODY")

TASK=$(printf '%s' "$SUBMIT_RESPONSE" | jq 'if has("data") then .data else . end')
PREDICTION_ID=$(printf '%s' "$TASK" | jq -r '.id')
if [ -z "$PREDICTION_ID" ] || [ "$PREDICTION_ID" = "null" ]; 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' "$RESPONSE" | jq 'if has("data") then .data else . end')
  STATUS=$(printf '%s' "$RESULT" | jq -r '.status')
  case "$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
Node.js example
const submitUrl = "https://api.wavespeed.ai/api/v3/wavespeed-ai/flux-2-klein-base-9b/edit-lora";
const apiKey = process.env.WAVESPEED_API_KEY;
if (!apiKey) throw new Error('Set WAVESPEED_API_KEY');

async function requestJson(url, options = {}) {
  const response = await fetch(url, options);
  if (!response.ok) throw new Error(await response.text());
  return response.json();
}

// 1. Submit the prediction.
const body = await requestJson(submitUrl, {
  method: "POST",
  headers: {
    "Authorization": `Bearer ${apiKey}`,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
        "prompt": "A cinematic shot of a city at sunset, soft golden light",
        "images": [
                "https://interactive-examples.mdn.mozilla.net/media/cc0-images/painted-hand-298-332.jpg"
        ],
        "seed": -1
}),
});
const task = body.data ?? body;
if (!task.id) throw new Error("Submission response did not contain a prediction id");
const resultUrl = task.urls?.get ||
  `https://api.wavespeed.ai/api/v3/predictions/${task.id}/result`;

// 2. Poll until the prediction finishes.
while (true) {
  const resultBody = await requestJson(resultUrl, {
    headers: { "Authorization": `Bearer ${apiKey}` },
  });
  const result = resultBody.data ?? resultBody;
  if (result.status === "completed") {
    console.log(result.outputs);
    break;
  }
  if (["failed", "cancelled", "timeout"].includes(result.status)) throw new Error(JSON.stringify(result));
  if (!["created", "processing"].includes(result.status)) throw new Error("Unexpected status: " + result.status);
  await new Promise(resolve => setTimeout(resolve, 2000));
}
Python example
import json
import os
import time
from urllib.request import Request, urlopen

api_key = os.environ["WAVESPEED_API_KEY"]
headers = {"Authorization": f"Bearer {api_key}", "Content-Type": "application/json"}
payload = {
    "prompt": "A cinematic shot of a city at sunset, soft golden light",
    "images": [
        "https://interactive-examples.mdn.mozilla.net/media/cc0-images/painted-hand-298-332.jpg"
    ],
    "loras": [],
    "seed": -1
}

def request_json(url, data=None):
    request = Request(url, data=data, headers=headers, method="POST" if data else "GET")
    with urlopen(request) as response:
        return json.load(response)

# 1. Submit the prediction.
body = request_json("https://api.wavespeed.ai/api/v3/wavespeed-ai/flux-2-klein-base-9b/edit-lora", json.dumps(payload).encode())
task = body.get("data", body)
if not task.get("id"):
    raise RuntimeError("Submission response did not contain a prediction id")
result_url = task.get("urls", {}).get("get") or f"https://api.wavespeed.ai/api/v3/predictions/{task['id']}/result"

# 2. Poll until the prediction finishes.
while True:
    result_body = request_json(result_url)
    result = result_body.get("data", result_body)
    status = result.get("status")
    if status == "completed":
        print(result.get("outputs", []))
        break
    if status in {"failed", "cancelled", "timeout"}:
        raise RuntimeError(result)
    if status not in {"created", "processing"}:
        raise RuntimeError(f"Unexpected status: {status}")
    time.sleep(2)

Flux 2 Klein Base 9b Edit Lora API — Frequently asked questions

What is the Flux 2 Klein Base 9b Edit Lora API?

Flux 2 Klein Base 9b Edit Lora is a WaveSpeedAI model for AI inference, exposed as a REST API on WaveSpeedAI. FLUX.2 [klein] Base 9B Edit with LoRA support is a high-quality image editing model with 9B parameters, offering precise modifications using natural language instructions and personalized styles via custom LoRA adapters. Ready-to-use REST inference API, best performance, no cold starts, affordable pricing. You can call it programmatically or try it from the playground above.

How do I call the Flux 2 Klein Base 9b Edit Lora API?

POST your input parameters to the model's REST endpoint (shown in the API tab of this playground) with your WaveSpeedAI API key in the Authorization header. Submission returns a prediction ID. Poll the result endpoint starting around every 2 seconds, increase the interval for long-running tasks, and stop on any terminal status. The playground generates production-oriented Python, JavaScript, and cURL examples with timeouts, transient-error handling, and safe GET retries. Full request/response shape is documented at https://wavespeed.ai/docs/docs-api/wavespeed-ai/flux-2-klein-base-9b-edit-lora.

How much does Flux 2 Klein Base 9b Edit Lora cost per run?

Flux 2 Klein Base 9b Edit Lora starts at $0.026 per run. That figure is the base price — the final charge scales with the parameters you set in the form (output size, length, count, references, or whatever knobs this model exposes), so a higher-quality or larger output costs more than a minimal one. The exact cost for your current input is shown live next to the Generate button before you submit, and the actual per-call charge is recorded on the prediction afterwards.

What inputs does Flux 2 Klein Base 9b Edit Lora accept?

Key inputs: `prompt`, `images`, `seed`, `enable_base64_output`, `enable_sync_mode`, `loras`. The full JSON schema (types, defaults, allowed values) is rendered above the Generate button and mirrored in the API reference at https://wavespeed.ai/docs/docs-api/wavespeed-ai/flux-2-klein-base-9b-edit-lora.

How do I get started with the Flux 2 Klein Base 9b Edit Lora API?

Sign up for a free WaveSpeedAI account to claim starter credits, copy your API key from /accesskey, then call the endpoint shown in the API tab of the playground. The playground also auto-generates a code sample in Python, JavaScript, or cURL for the parameters you've set.

Can I use Flux 2 Klein Base 9b Edit Lora outputs commercially?

Commercial usage rights depend on the model's license, set by its provider (WaveSpeedAI). The license summary appears on the model card above; see WaveSpeedAI's Terms of Service for platform-level conditions.