YuCollection
/

FLUX.2-klein-base-4B-Diffusers

@@ -14,13 +14,24 @@ tags:
 > **Note:** This repository is an **archived mirror** and is **not** the original upstream source.
 > The original model, weights, and documentation are developed and maintained by **Black Forest Labs**.
 >
-> All hosted model weights are **unmodified**. Any README edits are purely editorial and do not affect the model, its behavior, or its licensing.
 >
 > The model is released under the **Apache License, Version 2.0**.
 > A copy of the license is included in this repository.
 >
 > *This repository is not affiliated with or endorsed by Black Forest Labs.*
-> The maintainer is willing to cooperate in good faith with the original rights holder regarding reasonable requests.
 The **FLUX.2 [klein]** model family represents Black Forest Labs’ fastest image models to date. FLUX.2 [klein] unifies image generation and editing within a single compact architecture, **delivering state-of-the-art quality with end-to-end inference in under one second**. It is designed for real-time and latency-critical applications and runs on consumer hardware with as little as **~13GB VRAM**.
@@ -28,14 +39,6 @@ The **FLUX.2 [klein]** model family represents Black Forest Labs’ fastest imag
 It's a full-capacity foundation model. Undistilled, preserving complete training signal for maximum flexibility. Ideal for fine-tuning, LoRA training, research, and custom pipelines where control matters more than speed. Higher output diversity than the distilled models.
-**Official Announcement:**
-🔗 https://bfl.ai/blog/flux2-klein-towards-interactive-visual-intelligence
-## Official Upstream Sources
-- **black-forest-labs/FLUX.2-base-klein-4B (original Hugging Face model)**
-  https://huggingface.co/black-forest-labs/FLUX.2-klein-base-4B
 # **Key Features**
 1. Exceptional speed and quality-to-size ratio.

 > **Note:** This repository is an **archived mirror** and is **not** the original upstream source.
 > The original model, weights, and documentation are developed and maintained by **Black Forest Labs**.
 >
+> All hosted model weights are **unmodified**.
 >
 > The model is released under the **Apache License, Version 2.0**.
 > A copy of the license is included in this repository.
 >
 > *This repository is not affiliated with or endorsed by Black Forest Labs.*
+<p align="center">
+  <a href="https://bfl.ai/blog/flux2-klein-towards-interactive-visual-intelligence" target="_blank">
+    <img src="https://img.shields.io/badge/Announcement-Black%20Forest%20Labs-black?style=flat-square">
+  </a>
+  <a href="https://huggingface.co/black-forest-labs/FLUX.2-klein-base-4B" target="_blank">
+    <img src="https://img.shields.io/badge/Models-Hugging%20Face-yellow?style=flat-square">
+  </a>
+  <a href="https://github.com/black-forest-labs/flux2" target="_blank">
+    <img src="https://img.shields.io/badge/Source-GitHub-lightgrey?style=flat-square">
+  </a>
+</p>
 The **FLUX.2 [klein]** model family represents Black Forest Labs’ fastest image models to date. FLUX.2 [klein] unifies image generation and editing within a single compact architecture, **delivering state-of-the-art quality with end-to-end inference in under one second**. It is designed for real-time and latency-critical applications and runs on consumer hardware with as little as **~13GB VRAM**.
 It's a full-capacity foundation model. Undistilled, preserving complete training signal for maximum flexibility. Ideal for fine-tuning, LoRA training, research, and custom pipelines where control matters more than speed. Higher output diversity than the distilled models.
 # **Key Features**
 1. Exceptional speed and quality-to-size ratio.