FLUX.2-klein

Quantization Details 🔧

This model is a quantized version optimized for efficient inference:

• Transformer: Quantized using TorchAo fp8 (float8wo) quantization, significantly reducing model size while maintaining generation quality.
• Text Encoder: Replaced with unsloth/Qwen3-4B-unsloth-bnb-4bit, a 4-bit quantized version that further reduces memory requirements.
• Memory Usage: Peak VRAM consumption is approximately 9GB.
• Performance: Generates images in approximately 0.1 seconds (4 steps)on RTX 5090 GPUs.

Using with Diffusers 🧨

To use FLUX.2 [klein] 4B with the 🧨 Diffusers python library, first install or upgrade diffusers:

pip install -U diffusers

Then you can use Flux2KleinPipeline to run the model:

import os
import torch
from diffusers import Flux2KleinPipeline, Flux2Transformer2DModel

model_dir = "."
model_path = f"{model_dir}/FLUX.2-klein-4B-fp8-diffusers"
prompt = "A cat holding a sign that says hello Tonera"
height, width, guidance_scale, steps, seed = 1024, 1024, 4.0, 4, 0
dtype = torch.bfloat16

transformer = Flux2Transformer2DModel.from_pretrained(
    f"{model_path}/transformer",
    torch_dtype=dtype,
    use_safetensors=False,
)

pipe = Flux2KleinPipeline.from_pretrained(
    f"{model_path}",
    torch_dtype=dtype,
    transformer=transformer,
)
pipe.to("cuda")

img = pipe(
    prompt=prompt,
    height=height,
    width=width,
    guidance_scale=guidance_scale,
    num_inference_steps=steps,
    generator=torch.Generator(device="cuda").manual_seed(seed),
).images[0]

output = "output/flux2_klein.png"
os.makedirs(os.path.dirname(output) or ".", exist_ok=True)
img.save(output)
print(output)

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.
• Prompt following is heavily influenced by the prompting style.

Out-of-Scope Use

The model and its derivatives may not be used:

• In any way that violates applicable law.
• For the purpose of exploiting, harming or attempting to exploit or harm minors in any way; including but not limited to the solicitation, creation, acquisition, or dissemination of child exploitative content.
• To generate or disseminate deceptive, fraudulent, misleading or otherwise harmful content.
• To generate or disseminate personal identifiable information that can be used to harm an individual.
• To harass, abuse, threaten, stalk, or bully individuals or groups of individuals.
• To create non-consensual intimate imagery or illegal pornographic content.
• For fully automated decision making or high risk applications that adversely impact an individual's legal rights or otherwise create or modify a binding, enforceable obligation.

Nothing contained in this Model Card should be interpreted as or deemed a restriction or modification to the license the model is released under.

Hardware

The FLUX.2 [klein] 4B model fits in ~13GB VRAM and is accessible on NVIDIA RTX 3090/4070 and above.

Responsible AI Development

Black Forest Labs is committed to the responsible development and deployment of our models. Prior to releasing the FLUX.2 family of models, we evaluated and mitigated a number of risks in our model checkpoints and hosted services, including the generation of unlawful content, including child sexual abuse material (CSAM) and nonconsensual intimate imagery (NCII). We implemented a series of pre-release mitigations to help prevent misuse by third parties, with additional post-release mitigations to help address residual risks:

1. Pre-training mitigation. We filtered pre-training data for multiple categories of "not safe for work" (NSFW) and known child sexual abuse material (CSAM) to help prevent a user generating unlawful content in response to text prompts or uploaded images. We have partnered with the https://www.iwf.org.uk/, an independent nonprofit organization dedicated to preventing online abuse, to filter known CSAM from the training data.
2. Post-training mitigation. Subsequently, we undertook multiple rounds of targeted fine-tuning to provide additional mitigation against potential abuse, including both text-to-image (T2I) and image-to-image (I2I) attacks. By inhibiting certain behaviors and suppressing certain concepts in the trained model, these techniques can help to prevent a user generating synthetic CSAM or NCII from a text prompt, or transforming an uploaded image into synthetic CSAM or NCII.
3. Ongoing evaluation. Throughout this process, we conducted multiple internal and external third-party evaluations of model checkpoints to identify further opportunities for mitigation. External third-party evaluations focused on eliciting CSAM and NCII through adversarial testing with (i) text-only prompts, (