black-forest-labs/FLUX-1-dev
FLUX.1-dev is a state-of-the-art 12 billion parameter rectified flow transformer developed by Black Forest Labs. This model excels in text-to-image generation, providing highly accurate and detailed outputs. It is particularly well-regarded for its ability to follow complex prompts and generate anatomically accurate images, especially with challenging details like hands and faces.
FLUX.1-dev is a state-of-the-art 12 billion parameter rectified flow transformer developed by Black Forest Labs. This model excels in text-to-image generation, providing highly accurate and detailed outputs. It is particularly well-regarded for its ability to follow complex prompts and generate anatomically accurate images, especially with challenging details like hands and faces.
Input
text prompt
Num Images
number of images to generate (Default: 1, 1 ≤ num_images ≤ 4)
Num Inference Steps
number of denoising steps (Default: 25, 1 ≤ num_inference_steps ≤ 50)
image width in px (Default: 1024, 128 ≤ width ≤ 2048)
image height in px (Default: 1024, 128 ≤ height ≤ 2048)
random seed, empty means random (Default: empty, 0 ≤ seed < 18446744073709552000)
Guidance Scale
classifier-free guidance, higher means follow prompt more closely (Default: 7.5, 0 ≤ guidance_scale ≤ 20)
You need to login to use this model
Output
FLUX.1-dev
FLUX.1-dev is a 12 billion parameter rectified flow transformer developed by Black Forest Labs. It is designed for high-quality text-to-image generation and offers competitive performance comparable to closed-source models. The model leverages latent adversarial diffusion distillation, allowing for efficient image generation in just 25 to 50 steps. It is available under a non-commercial license for personal, scientific, and research purposes.
Key Features
- Cutting-Edge Output Quality: FLUX.1-dev delivers images that are both detailed and aligned with complex prompts, providing quality on par with leading proprietary models.
- Efficient Training: The model's training using latent adversarial diffusion distillation enables the generation of high-quality images in 25 to 50 steps, optimizing both speed and accuracy.
- Open-Source for Non-Commercial Use: FLUX.1-dev can be freely used for non-commercial purposes, making it accessible to researchers, artists, and developers.
- Versatile Style Range: It supports a wide range of visual styles, from photorealistic to surreal artistic creations, suitable for various creative tasks.
Limitations
- Factual Limitations: As a statistical image generation model, FLUX.1-dev is not designed to provide factual information and may produce images that are disconnected from real-world accuracy.
- Bias Concerns: Like many AI models, FLUX.1-dev can amplify existing societal biases present in its training data.
- Prompt Dependency: The model's performance is closely tied to the clarity and structure of the prompts. For the best results, prompts may need to be carefully crafted and fine-tuned.
Out-of-Scope Use
- Legal Violations: The model cannot be used in any manner that violates local, national, or international laws and regulations.
- Harmful Content: It is strictly prohibited to use FLUX.1-dev for harmful purposes, such as the creation of exploitative content, harassment, or disinformation.
- Non-Consensual Content: The generation of non-consensual nudity or illegal pornographic content is forbidden, as is using the model for automated decision-making that infringes on individuals' rights.
