FLUX.2-klein-4B by Black Forest Labs: Open-Weight Model

Runs On

What it takes to serve FLUX.2-klein-4B (3.9B parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.

Precision	Weights	Memory needed	Cheapest setup	Per hour	Also fits
16-bit	7.8 GB	9.3 GB	1x MI300X (192 GB) Vultr	$1.85	1x H100 $1.99 · 1x MI325X $2.00
8-bit	3.9 GB	4.7 GB	1x MI300X (192 GB) Vultr	$1.85	1x H100 $1.99 · 1x MI325X $2.00
4-bit	1.9 GB	2.3 GB	1x MI300X (192 GB) Vultr	$1.85	1x H100 $1.99 · 1x MI325X $2.00

Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the SAVRN Index, read Sep 18, 2026.

SAVRN's Notes on FLUX.2-klein-4B

Nine point three gigabytes of memory at 16-bit, 7.8 GB of that weights. That settles hardware for FLUX.2-klein-4B: a 3.9B parameter rectified flow transformer that generates images from text and edits them against multiple references does not need a 192 GB MI300X. That card, the cheapest in our table at $1.85 an hour on demand, gets shared across instances or swapped for smaller silicon. At 8-bit the need is 4.7 GB, at 4-bit 2.3 GB, and the publisher states a 13 GB VRAM floor.

Under Apache 2.0 you can use it commercially, modify it and redistribute it, keeping the license, copyright notices and any NOTICE file, stating significant changes, with an express patent grant. No context length is listed and no base model or training dataset is named, so provenance is what the publisher supplies, and with a 2026-01-14 release updated 2026-02-24, pin the version you validated.

SAVRN Research, 2026-09-18

Model Card

By Black Forest Labs, published under apache-2.0, revision e7b7dc27f91d.

The FLUX.2 [klein] model family are our fastest image models to date. FLUX.2 [klein] unifies generation and editing in a single compact architecture, delivering state-of-the-art quality with end-to-end inference in as low as under a second. Built for applications that require real-time image generation without sacrificing quality, and runs on consumer hardware, with as little as 13GB VRAM.

FLUX.2 [klein] 4B is a 4 billion parameter rectified flow transformer capable of generating images from text descriptions and supports multi-reference editing capabilities.

Fully open under Apache 2.0. Our most accessible model runs on consumer GPUs like the RTX 3090/4070. Compact but capable: supports text-to-image, image editing, and multi-reference at quality that punches above its size. Built for local development, edge deployment, and production use.

For more information, please read our blog post.

Key Features

Read the full model card (1,307 words)

Identity and Version

Repository: black-forest-labs/FLUX.2-klein-4B
Publisher: Black Forest Labs
Task: Image to image
Modality: Image
Library: diffusers
Parameters: 3.9B parameters
Languages: en
Revision: e7b7dc27f91deacad38e78976d1f2b499d76a294
First published: 2026-01-14
Last updated: 2026-02-24

Files and Weights

25 files, 23.7 GB in total. The weights are 5 files totalling 23.7 GB in safetensors.

Weights5 files · 23.7 GB

Configuration9 files · 38.2 KB

Tokenizer4 files · 15.9 MB

Documentation2 files · 19.6 KB

Other4 files · 8.8 MB

Repository1 file · 1.6 KB

Every file

File	Type	Size	SHA-256
flux-2-klein-4b.safetensors	Weights	7.8 GB	ec3d4e733a77
text_encoder/model-00001-of-00002.safetensors	Weights	5.0 GB	8c0506e7f493
text_encoder/model-00002-of-00002.safetensors	Weights	3.1 GB	82f2bd839378
transformer/diffusion_pytorch_model.safetensors	Weights	7.8 GB	9f29f9edcfda
vae/diffusion_pytorch_model.safetensors	Weights	168.1 MB	ca70d2202afe
model_index.json	Configuration	446 B	—
scheduler/scheduler_config.json	Configuration	486 B	—
text_encoder/config.json	Configuration	1.5 KB	—
text_encoder/generation_config.json	Configuration	214 B	—
text_encoder/model.safetensors.index.json	Configuration	32.9 KB	—
tokenizer/added_tokens.json	Configuration	707 B	—
tokenizer/special_tokens_map.json	Configuration	613 B	—
transformer/config.json	Configuration	541 B	—
vae/config.json	Configuration	821 B	—
LICENSE.md	Documentation	9.6 KB	—
README.md	Documentation	10.0 KB	—
editing.jpg	Other	2.5 MB	2912ca5a7cb9
others.jpg	Other	3.4 MB	78caed191515
realism.jpg	Other	2.9 MB	6cccc69a683c
tokenizer/chat_template.jinja	Other	4.2 KB	—
.gitattributes	Repository	1.6 KB	—
tokenizer/merges.txt	Tokenizer	1.7 MB	—
tokenizer/tokenizer.json	Tokenizer	11.4 MB	aeb13307a71a
tokenizer/tokenizer_config.json	Tokenizer	5.4 KB	—
tokenizer/vocab.json	Tokenizer	2.8 MB	—

License and Download

License: apache-2.0
Access: Open weights, no gate
Download size: 23.7 GB

Download from Black Forest Labs

Released by Black Forest Labs through its official repository on Hugging Face. Read the license.

Memory Requirements

Precision	Weights in memory
As published	23.7 GB
16-bit	7.8 GB
8-bit	3.9 GB
4-bit	1.9 GB

Weights only, from the published parameter count; the key-value cache and runtime add to this.

Built on This Model

Quantized fromFLUX.2-klein-4B-GGUF
Derived fromFLUX.2-klein-4B-GGUF
Derived fromFLUX.2-klein-4B-mflux-4bit

Questions About FLUX.2-klein-4B

How much GPU memory does FLUX.2-klein-4B need?

About 9.3 GB at 16-bit and 2.3 GB at 4-bit: the weights (3.9B parameters) plus a working margin. A long context needs more.

What is the cheapest GPU to run FLUX.2-klein-4B on?

At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.

Can I use FLUX.2-klein-4B commercially?

Yes. FLUX.2-klein-4B is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.

Similar Models

Model · Image to image

FLUX.2-klein-base-4B

Black Forest Labs

The FLUX.2 [klein] model family are our fastest image models to date. FLUX.2 [klein] unifies generation and editing in a single compact architecture, delivering state-of-the-art quality with end-to-end inference in as low as under a second. Built for applications that require real-time image generation without sacrificing quality, and runs on consumer hardware, with as little as 13GB VRAM. FLUX.2 [klein] 4B Base is a 4 billion parameter rectified flow transformer capable of generating images from text descriptions and supports multi-reference editing capabilities. It's a full-capacity foundation model. Undistilled, preserving complete training signal for maximum flexibility. Ideal for…

Open weights apache-2.0 3.9B parameters diffusers

View model

Model · Image to image

FLUX.2-small-decoder

Black Forest Labs

FLUX.2 Small Decoder is a distilled VAE decoder that serves as a drop-in replacement for the standard FLUX.2 decoder. It delivers faster decoding and lower VRAM usage with minimal to zero quality loss. The encoder remains unchanged. 1. ~1.4x faster decoding compared to the full decoder. 2. ~1.4x less VRAM at decode time, enabling higher resolutions without running out of memory. 3. ~28M decoder parameters (vs ~50M in the full decoder) thanks to narrower channel widths ([96, 192, 384, 384] vs [128, 256, 512, 512]). 4. Minimal quality loss — images are almost identical. 5. Available under the Apache 2.0 license. Compatible with all open FLUX.2 models: - This model is not intended or able to…

Open weights apache-2.0 62M parameters diffusers

View model

Model · Image to image

FLUX.2-klein-4B-GGUF

Unsloth AI

This is a GGUF quantized version of FLUX.2-klein-4B. unsloth/FLUX.2-klein-4B-GGUF uses Unsloth Dynamic 2.0 methodology for SOTA performance. - Important layers are upcasted to higher precision. - Uses tooling from ComfyUI-GGUF by city96. The FLUX.2 [klein] model family are our fastest image models to date. FLUX.2 [klein] unifies generation and editing in a single compact architecture, delivering state-of-the-art quality with end-to-end inference in as low as under a second. Built for applications that require real-time image generation without sacrificing quality, and runs on consumer hardware, with as little as 13GB VRAM. FLUX.2 [klein] 4B is a 4 billion parameter rectified flow…

Open weights apache-2.0 ggml

View model

Model · Image to image

FLUX.2-klein-4b-fp8

Black Forest Labs

FLUX.2 [klein] 4B is a 4 billion parameter rectified flow transformer capable of generating images from text descriptions and supports multi-reference editing capabilities. For more information, please read our blog post. This repository holds an FP8 version of FLUX.2 [klein] 4B. The main repository of this model (full BF16 weights) can be found here. Limitations - This model is not intended or able to provide factual information. - While the model can output text, text rendered may be inaccurate or subject to distortion. - As a statistical model, this checkpoint may represent or amplify biases observed in the training data. - The model may fail to generate output that matches the prompts.…

Open weights apache-2.0 diffusion-single-file

View model

Model · Image to image

qwen-image-edit-plus-nsfw-lora

ScottzillaSystems

A LoRA adapter for the Qwen-Image-Edit-2511 diffusion pipeline, enabling NSFW content generation and editing capabilities. The easiest way to use this LoRA is through the ScottzillaSystems Image Editor — select " MCNL-NSFW-v1" from the adapter dropdown. Use these concepts in your prompts to activate specific capabilities: nsfw nipples vagina penis missionary cowgirlout reversecowgirlpov blowjob cumonface creamp1e l1ck

Open weights openrail++ diffusers

View model

Model · Image to image

FLUX.2-klein-base-4b-fp8

Black Forest Labs

FLUX.2 [klein] 4B Base is a 4 billion parameter rectified flow transformer capable of generating images from text descriptions and supports multi-reference editing capabilities. For more information, please read our blog post. This repository holds an FP8 version of FLUX.2 [klein] 4B Base. The main repository of this model (full BF16 weights) can be found here. Limitations - This model is not intended or able to provide factual information. - While the model can output text, text rendered may be inaccurate or subject to distortion. - As a statistical model, this checkpoint may represent or amplify biases observed in the training data. - The model may fail to generate output that matches the…

Open weights apache-2.0 diffusion-single-file

View model