- AI 模型目录
- FLUX Pro
FLUX Pro
Black Forest Labs
参数配置
生成结果

样例展示

A hyperrealistic portrait of an elderly fisherman with weathered skin, stormy sea background, dramatic Rembrandt lighting, 8K photography
规格参数
定价
credits per image
~15 credits for a 1024x1024 image
FLUX Pro is the flagship image generation model from Black Forest Labs, the team behind the original Stable Diffusion research. Built on a novel rectified flow transformer, FLUX Pro sets a new bar for prompt adherence — the degree to which every detail specified in a text prompt is faithfully rendered in the output image. Whether you are producing product photography, editorial portraits, or concept art, FLUX Pro delivers photorealistic output at up to 2048x2048 resolution with commercial licensing included.
能力说明
FLUX Pro outperforms earlier diffusion models on three critical dimensions: prompt adherence (complex multi-subject scenes with precise spatial relationships), texture fidelity (skin pores, fabric weave, metallic surfaces, and natural materials), and lighting accuracy (directional shadows, subsurface scattering, and specular highlights). The model handles five aspect ratios natively without quality degradation, making it suitable for a wide range of production formats from vertical social content to cinematic widescreen compositions. Commercial licensing is included by default.
最佳实践
FLUX Pro performs best when prompts include specific lighting descriptors, material properties, and camera characteristics. For example: "a glass perfume bottle on a white marble surface, studio softbox lighting, macro lens, bokeh background, commercial product photography" will outperform a vague prompt like "a perfume bottle". When working on portraits, specify lens focal length (e.g. "85mm portrait lens") and lighting setup ("Rembrandt lighting", "golden hour backlight") for the most photorealistic results. Use negative space descriptions sparingly — FLUX Pro is better guided by what you want than by what you want to exclude.
常见问题
What resolution does FLUX Pro support?
FLUX Pro generates images from 512x512 up to 2048x2048 pixels. The model natively supports five aspect ratios (1:1, 16:9, 9:16, 3:2, 2:3), so you can produce portrait, landscape, and square images without cropping or letterboxing artefacts.
How does FLUX Pro achieve such strong prompt adherence?
FLUX Pro is based on a rectified flow transformer architecture trained on a highly curated dataset with rigorous caption quality controls. This allows the model to map text tokens to visual regions far more precisely than earlier UNet-based diffusion models, minimising hallucinated elements and correctly placing subjects, attributes, and spatial relationships described in the prompt.
Can FLUX Pro images be used in commercial products?
Yes. Black Forest Labs grants a broad commercial licence for images generated via the FLUX Pro API. You can use outputs in advertising, product packaging, editorial publications, and software products without additional royalty payments, subject to Genace and Black Forest Labs terms of service.
相关模型

Ideogram v3
Ideogram
Create logos, posters, and typographic designs with perfect text rendering using Ideogram v3. The only image AI that consistently generates legible, stylised text inside images — ideal for brand identity, marketing, and editorial design.

Nano Banana
Nano Banana is the official nickname for Google Gemini 2.5 Flash Image. Generate and iteratively edit images through natural language conversation, maintaining character and scene consistency across multiple turns — all at impressive speed via the Genace API.
FLUX 2
Black Forest Labs
FLUX 2 by Black Forest Labs is the next generation of the acclaimed FLUX architecture, promising native 4K image generation, faster inference, and even stronger prompt adherence than FLUX Pro. Join the Genace waitlist to access it at launch.