muapi Logomuapi Logo
muapi Logomuapi Logo

FLUX.2 Klein 4B Edit LoRA: Lora Support

Flux-2-Klein-4B Edit with LoRA performs instruction-based image edits while applying custom LoRA adapters, letting you preserve scene layout and lighting while pushing the result toward a specific style or character look. Great for stylized object swaps and look-conditioned edits.

Interactive model controls

Flux-2-Klein-4B Edit with LoRA performs instruction-based image edits while applying custom LoRA adapters, letting you preserve scene layout and lighting while pushing the result toward a specific style or character look. Great for stylized object swaps and look-conditioned edits.

📝

Overview

About this model

Flux-2-Klein-4B Edit with LoRA performs instruction-based image edits while applying custom LoRA adapters. It preserves the original scene's layout and lighting while pushing the result toward a specific style, character, or look defined by your LoRA. This makes it well-suited for stylized object swaps, costume changes, and look-conditioned edits where consistency with a brand or character LoRA matters.

You can pass 1 to 3 reference images plus an editing prompt, and stack up to 3 LoRA adapters per request. The 4B base keeps cost and latency low, which is ideal for high-volume editorial pipelines and rapid creative iteration.

1Stylized object swaps and costume changes
2Look-conditioned product edits aligned with a brand LoRA
3Character-consistent edits using a character LoRA
4Themed enhancements (e.g. anime / illustration overlays) on real photos
5Bulk editorial restyling at low per-image cost
💰

Pricing & Value

Cost analysis

muapiapp$0.025 per generation

Flat per-image price for the 4B Edit-with-LoRA variant, regardless of LoRA count.

Fal.aiNot available

Flux-2 Klein Edit with LoRA is not generally available on Fal.ai at the time of writing.

ReplicateNot available

Flux-2 Klein Edit with LoRA is not generally available on Replicate at the time of writing.

** Competitor pricing is estimated based on similar model architectures and usage tiers.

⚙️

Technical Details

Configuration schema

Promptstring

Editing instruction describing the desired change.

Default ValueAdd a tiny blue knitted scarf around the kitten's neck, keep pose, lighting, and background unchanged.
Image URLsarray

List of 1-3 reference image URLs to edit.

Default Valuehttps://d3adwkbyhxyrtq.cloudfront.net/webassets/videomodels/flux-2-klein-4b-edit-in.jpg
LoRA Listarray

Up to 3 LoRA adapters to apply during the edit.

Default Value[object Object]
Aspect RatioEnum (7 options)

The aspect ratio of the generated image

Default Value1:1
📖

Implementation Guide

Developer documentation

How to Use Flux-2-Klein-4B Edit with LoRA

  1. Provide reference images: Pass 1 to 3 image URLs in images_list to edit.

  2. Write an editing instruction: Describe the change you want in prompt. Keep the change targeted so the original layout and lighting are preserved.

  3. Attach LoRAs: Provide up to 3 entries in lora_list, each with a path and scale.

  4. Choose an aspect ratio: Use aspect_ratio to control output framing.

  5. Submit and iterate: Call flux-2-klein-4b-edit-lora. Adjust LoRA scale or sharpen the prompt if the edit is too weak or too strong and re-run.

❓

Common Questions

Frequently asked

How many reference images can I provide?

Between 1 and 3 reference images per edit request, passed in `images_list`.

Will the edit preserve identity / pose?

Klein Edit aims to preserve the original scene's layout and lighting. Identity preservation is best when your prompt targets specific elements (clothing, background, accessories) rather than rebuilding the subject.

Can I use the same LoRA workflow for both edit and text-to-image?

Yes. The `lora_list` schema is identical across the text-to-image and edit endpoints, so the same adapter weights and `scale` settings transfer cleanly between them.