Flux.2-Klein-9B-MatchingPose

Flux.2-Klein-9B-MatchingPose

A LoRA for FLUX.2 Klein 9B that puts a character into the pose shown by a mannequin reference while keeping the character's face and style. For artists and designers. Use only faces you have the right to use.

Made by nhathoangfoto on Hugging Face. These notes are Mindket's, written from its model page and model card; the model's own files and license are on Hugging Face.

About

What it is: Flux.2-Klein-9B-MatchingPose is a LoRA adapter for the FLUX.2 Klein 9B image model. It takes a mannequin pose reference and a character reference, and produces the character in that exact pose while keeping its face, identity and style. It is the second stage of a pose-transfer pipeline, and it pairs with a companion mannequin LoRA by the same author.

Who it suits: It suits artists and designers who need the same character in many poses without losing the face from one image to the next. Use only faces and characters you have the right to use.

Industry | Who uses it | How they use it | Example
Illustration | Character artists | Put one character into a planned pose | A character face placed into a mannequin pose
Games and animation | Designers | Keep a character consistent across poses | One face reused across several poses
Marketing | Content creators | Create posed character visuals | A branded mascot posed to match a layout
Research | Image generation tinkerers | Study pose transfer without ControlNet drift | Compare it with a text-only pose prompt

If you need a general image generator with no pose reference, you can skip it.

Where to use it: It runs with diffusers on top of the FLUX.2 Klein 9B base model. Its files download to Windows, macOS or Linux.

• Python with the diffusers library
• The FLUX.2 Klein 9B base model it was built on
• Windows, macOS or Linux with the huggingface_hub package

Use case | Who | What they get back
Pose transfer | Artists | The character repeating the mannequin's limb positions and body direction
Character consistency | Designers | The same face, identity and style across images
Two-stage pipelines | Creators | A pose template from a photo, then a posed character

How to set it up: You need Python, the diffusers library, the base model and this LoRA. The card's instructions for running it are not shown in the part available here.

• Python on Windows, macOS or Linux
• The Hugging Face hub package
• The FLUX.2 Klein 9B base model

1. Install the Hugging Face hub package.
2. Download the LoRA files.
3. Load them on top of the base model, following the model card.

pip install -U huggingface_hub
hf download nhathoangfoto/Flux.2-Klein-9B-MatchingPose

How to use it: Give the model two images, a face or character reference and a mannequin pose reference, and it renders the character in that pose. The mannequin can come from the companion Mannequin LoRA, which turns any photo into a faceless pose template.

Tool | What it does
Mannequin LoRA, stage 1 | Turns a real photo into a faceless mannequin pose template
MatchingPose LoRA, stage 2 | Places your character into that pose, as this model does
Character reference | Supplies the face, identity and style to keep

Example: You supply a character face and a mannequin pose that shows a complex stance. The model returns your character in that stance, with natural skin, clothing and lighting, not a stiff mannequin render.

For newcomers and experts: If you are new to this, picture a posable wooden figure that sets the pose while a second image sets who the character is. A first small step is to try the card's simple base pose before a complex one.

If you know the field, the mannequin works as an identity-free pose anchor, so the character LoRA can supply identity without pose drift. The card says common alternatives either change the character's face, fail to hit the pose from text alone, or need several ControlNets at once.

Real reference photo
Mannequin LoRA makes a pose template
Character reference supplies the face
MatchingPose LoRA joins both
Final image in the same pose

Good to know: The card says it works with any input face, including photographs, AI-generated characters and illustrations. Its output follows the mannequin's limbs, body direction and framing.

• License: Apache-2.0
• Cost to run: no API key or paid service is named, and the card gives no hardware figures in the part shown
• Data it reaches: the download comes from Hugging Face, and the model works on the face and pose images you give it
• Not for: using a face or likeness you have no right to use

On Hugging Face, 13 November 2026

• Task: image text to image
• Library: diffusers
• Downloads in the last 30 days: 3,319
• Likes: 49
• License: apache-2.0
• Languages: en
• Based on: black-forest-labs/FLUX.2-klein-base-9B
• Tags: diffusers, lora, flux, flux.2, flux.2-klein, matchingpose, pose-transfer, character-consistency, image-to-image, text-to-image, image-text-to-image

Links

• Model: https://huggingface.co/nhathoangfoto/Flux.2-Klein-9B-MatchingPose
• Files: https://huggingface.co/nhathoangfoto/Flux.2-Klein-9B-MatchingPose/tree/main

Get the model

Download its files with Hugging Face's command line:

pip install -U huggingface_hub
hf download nhathoangfoto/Flux.2-Klein-9B-MatchingPose

Then follow the usage in its model card. Nothing in this kit runs by itself.