
Max-quality image generation and editing with support for ten reference images
Frequently asked questions
What is Flux 2 Flex and who made it?
Flux 2 Flex (model ID: black-forest-labs/flux-2-flex) is a flagship AI image generation and editing model created by Black Forest Labs. It is available on CRAISEE via Replicate and is designed for professional-grade creative production, offering maximum output quality with extensive control over style, composition, and prompt adherence.
How many reference images can Flux 2 Flex accept at once?
Flux 2 Flex supports up to ten simultaneous reference images. This multi-reference capability makes it ideal for complex compositional work, style blending, and guided image generation where multiple visual sources need to be combined or referenced in a single output.
What image resolution does Flux 2 Flex support?
Flux 2 Flex generates images up to 4 megapixels, with a maximum resolution of 2048×2048 pixels. For the best balance between quality and performance, 2 MP is the recommended output size. It also supports flexible aspect ratios including 1:1, 16:9, 9:16, 3:2, 4:3, and custom dimensions.
What output formats does Flux 2 Flex export?
Flux 2 Flex supports three output formats: WebP, JPEG, and PNG. This flexibility ensures compatibility with a wide range of downstream production pipelines, whether you need compressed web-ready images or lossless files for professional post-production workflows.
Can Flux 2 Flex edit existing images?
Yes. Flux 2 Flex includes image-to-image editing capabilities, allowing users to modify, extend, or reinterpret existing images using text prompts combined with reference inputs. This makes it suitable for retouching, style transfer, and creative reinterpretation of source material.
How popular is Flux 2 Flex on CRAISEE?
Flux 2 Flex has surpassed 500,000 runs on CRAISEE, making it one of the platform's most widely used and trusted models. Its popularity reflects its reliability for professional-grade image generation and editing tasks across a broad range of creative use cases.
What makes Flux 2 Flex different from other text-to-image models?
Flux 2 Flex stands out through its combination of high-fidelity text rendering, support for up to ten reference images, adjustable inference steps for quality-speed trade-offs, and guidance scale control for prompt adherence. Few models offer this level of compositional flexibility alongside maximum-quality output in a single unified tool.
How do I control quality vs. speed in Flux 2 Flex?
Flux 2 Flex lets users adjust the number of inference steps to balance speed and quality. Fewer steps enable fast iteration during early creative stages, while more steps produce maximum-quality final renders. The guidance scale parameter further controls how strictly the model follows the text prompt versus interpreting it more expressively.



