Black Forest Labs releases FLUX 3 Image Klein, a 5.6B open weights model

What the model offers

FLUX 3 Image Klein shares core capabilities with the full FLUX 3 Image model, according to Black Forest Labs: native 4K generation, pixel-perfect editing, and bounding boxes that control where elements are placed or confine edits to a defined area.

The company says the model builds on shared learnings of the FLUX 3 family across images, video and audio. Images contribute spatial detail, video adds motion and physical dynamics, and audio provides information about events and interactions. Together, these signals are meant to help the model generate images that feel grounded in the real world.

The release includes a base model for fine-tuning and a step distilled model for low-latency generation. FP8, INT8 and NVFP4 builds were developed with NVIDIA support. All are published under the FLUX Kommunity License.

Quality and speed claims

In internal evaluations, Black Forest Labs says FLUX 3 Image Klein ranks first in text-to-image, multi-reference generation and image editing, ahead of larger open weights models. It also rates above the closed models in the comparison, Ideogram V4.5 and FLUX.2 [pro].

At 5.6B parameters, it is the smallest open model in the comparison, against 7B for Qwen-Image 2.1 and 12.9B for Krea 2. Compared with FLUX.2 [klein] 9B, it gains 91 Elo in text-to-image and 70 in image editing, with 38% fewer parameters.

Through the BFL API, the company says the model delivers lower latency than Qwen-Image 2.1, FLUX.2 [pro], Ideogram V4.5 and Krea 2, with the highest rating at its latency and above. Black Forest Labs notes that its model was timed on its own servers, while other models were timed from the client, which includes polling and network time. Krea 2 was timed on its Medium tier, and its Large tier took 31.1 seconds for text-to-image.

Built for consumer GPUs

The model was optimized for consumer hardware. The generator weights require 11.2 GB in BF16 and 5.6 GB in FP8. With the INT8 build, it runs on NVIDIA GeForce RTX GPUs with as little as 8GB of VRAM, through ComfyUI or Diffusers.

With the distilled model and the INT8 build, it generates a 1MP image in about 4 seconds on an NVIDIA GeForce RTX 5090 and under 9 seconds on an RTX 4070 through ComfyUI, according to measurements by NVIDIA. On an NVIDIA GB300, the distilled model generates a 1MP image in under 0.6 seconds in NVFP4. Serving a LoRA adds around 5% to generation time.

Black Forest Labs argues that fewer parameters make the model practical to run locally and cheaper to adapt. Teams with data sovereignty or local deployment requirements can evaluate and adapt it on their own infrastructure, and the company points to on-device image editing on laptops as a key use case.

Availability, license and safety

The distilled model is available in the BFL API as flux-3-image-klein, with prompt upsampling and pixel restoration, which preserves unedited regions of an image exactly. Pricing starts at $0.009 per 1K image. The weights for the base and distilled models are on Hugging Face, along with reference inference code and a Gradio demo for bounding-box control.

Day-one support covers ComfyUI, Hugging Face Diffusers, TensorRT-LLM, SGLang Diffusion and vLLM-Omni. For fine-tuning, Diffusers and Ostris's AI Toolkit support LoRA training from day one, and NVIDIA's NeMo AutoModel ships recipes that work directly from the released checkpoints.

The FLUX Kommunity License allows free commercial use for individuals and companies with annual revenue under $5M, alongside research and non-commercial use. Larger organizations can deploy commercially through self-hosted licensing options or the API.

On safety, the company cites dataset filters during pre-training and targeted fine-tuning during post-training. The checkpoints were red teamed internally and evaluated by external safety partner Cinder, which the company says showed substantially fewer vulnerabilities than other popular open weights models.

What it means

For developers and teams, the release offers an open weights image model small enough for consumer GPUs, with editing, multi-reference input and fine-tuning support from launch. Black Forest Labs says FLUX 3 Image Klein is the second open weights release from the FLUX 3 family, following FLUX 3 Action, and that more will follow. The quality and speed rankings come from the company's own evaluations.