1. Flux.2 Series
- Flux.2 dev
- Flux.2 kelin 4b
- Flux.2 kelin 9b
- Flux.2 4b base
- Flux.2 9b base
Multi-Reference Support: Reference up to 10 images simultaneously with the best character / product / style consistency available today.
- Image Detail & Photorealism: Greater detail, sharper textures, and more stable lighting suitable for product shots, visualization, and photography-like use cases.
- Text Rendering: Complex typography, infographics, memes and UI mockups with legible fine text now work reliably in production.
- Enhanced Prompt Following: Improved adherence to complex, structured instructions, including multi-part prompts and compositional constraints.
- World Knowledge: Significantly more grounded in real-world knowledge, lighting, and spatial logic, resulting in more coherent scenes with expected behavior.
- Higher Resolution & Flexible Input/Output Ratios: Image editing on resolutions up to 4MP.
Flux.2 dev
32B open-weight model, derived from the FLUX.2 base model. The most powerful open-weight image generation and editing model available today, combining text-to-image synthesis and image editing with multiple input images in a single checkpoint.
Download counts: 920k
Model size: 64GB
Adapter
- flux.2 dev turbo
description: FLUX.2 dev Turbo is a distilled LoRA adapter for FLUX.2 dev that enables high-quality image generation in just 8 inference steps.
- berthe-morisot lora
Model link
tips: NVFP is a model suitable for nvidia blackwell GPU architecture (rtx 50 series)
Flux.2 kelin 4B
The FLUX.2 klein model family are the 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.
dowmload counts: 200k
Model size: 4.07GB
Requirements at least 13GB VRAM
adapters
- spritesheet-lora
description: Transform any object into a 2×2 multi-view sprite sheet with consistent style
prompts: 2x2 sprite sheet
- outpainting-lora
description: This LoRA fills green border areas around an image with contextually appropriate AI-generated content. Simply add a green border to any image, and the model will extend the scene naturally.
prompts: Fill the green spaces according to the image
Model link
Flux.2 kelin 9B
FLUX.2 klein 9B is a 9 billion parameter rectified flow transformer capable of generating images from text descriptions and supports multi-reference editing capabilities.
Our flagship small model. Defines the Pareto frontier for quality vs. latency across text-to-image, single-reference editing, and multi-reference generation. Matches or exceeds models 5x its size—in under half a second. Built on a 9B flow model with 8B Qwen3 text embedder, step-distilled to 4 inference steps.
Dowmload counts: 100k
Model size: 18.2GB
Adapters
- consistency-lora(image-image)
- Anything to real characters-lora(text-image)
description: This LoRA is designed for Anime-to-Realism (Live-Action) conversion, specifically optimized for the Flux2 Klein architecture.
prompts: realistic style of a young Asian girl
Model link:
Flux.2 4b base
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.
Download counts: 90k
Model size: 7.75GB
Adapter
- pixel spitesheet 4walk lora
description: A pixel art spritesheet LoRA for small 32x32 characters, with animation frames for walking up/down/left/right, standing with both arms raised, jumping left/right, and lying on the floor.
Model link:
Flux.2 9b base
FLUX.2 klein 9B Base is a 9 billion parameter rectified flow transformer capable of generating images from text descriptions and supports multi-reference editing capabilities.
Download counts: 75k
Model size: 9.57GB
Adapter
- Models-lora
Description:
Model link:
Some notes for beginners: fp8 version only includes diffusion model. Vae, text_encoders and other models are not included in the fp8 version. If you need them, enter the original model page to download them. If you don't know what to do, go to download models tutorial