flux2-klein-9B
Back
flux2-klein-9B
flux2-klein-9B

flux2-klein-9B

428.8K Views3 Likes161 Favorites
Other

初级A

初级A

Other

Model Information

Active
Original author:
转载
Model Type:
Unet
Basic Model:
F2基础
Trigger Words:
无
Resource Name:
models/unet/flux-2-klein-9b-nvfp4.safetensors
MD5:
f6480b7c68d6af6cefcfd1490d2619b7

FLUX.2 [klein] model series is the fastest image model we have developed to date. FLUX.2 [klein] integrates generation and editing into a compact architecture, providing state-of-the-art quality and end-to-end inference times as low as under one second. Designed for applications requiring real-time image generation without compromising quality.
FLUX.2 [klein] 9B is a rectified flow transformer with 9 billion parameters, capable of generating images from text descriptions and supporting multi-reference editing functionality.
Our flagship compact model. It defines the Pareto frontier of quality and latency for text-to-image, single-reference editing, and multi-reference generation. In less than half a second, it matches or exceeds the performance of models five times its size. Based on the 9B flow model, equipped with an 8B Qwen3 text embedder, and refined into 4 inference steps.
This repository hosts the NVFP4 version of FLUX.2 [klein] 9B. The primary repository for this model (complete BF16 weights) can be found here.

This model is sourced from an external transfer (transfer address: https://modelscope.cn/models/black-forest-labs/FLUX.2-klein-9b-nvfp4 ),if the original author has objections to this transfer, you can click,
Appeal
We will, within 24 hours, edit, delete, or transfer the model to the original author according to the original author's request

FLUX.2 [klein] model series is the fastest image model we have developed to date. FLUX.2 [klein] integrates generation and editing into a compact architecture, providing state-of-the-art quality and end-to-end inference times as low as under one second. Designed for applications requiring real-time image generation without compromising quality.
FLUX.2 [klein] 9B is a rectified flow transformer with 9 billion parameters, capable of generating images from text descriptions and supporting multi-reference editing functionality.
Our flagship compact model. It defines the Pareto frontier of quality and latency for text-to-image, single-reference editing, and multi-reference generation. In less than half a second, it matches or exceeds the performance of models five times its size. Based on the 9B flow model, equipped with an 8B Qwen3 text embedder, and refined into 4 inference steps.
This repository hosts the NVFP4 version of FLUX.2 [klein] 9B. The primary repository for this model (complete BF16 weights) can be found here.

This model is sourced from an external transfer (transfer address: https://modelscope.cn/models/black-forest-labs/FLUX.2-klein-9b-nvfp4 ),if the original author has objections to this transfer, you can click,
Appeal
We will, within 24 hours, edit, delete, or transfer the model to the original author according to the original author's request