The premier open text-to-image model family with photorealistic 4MP output
FLUX.2 represents the next generation of Black Forest Labs' groundbreaking FLUX model family, which has become the dominant open-source architecture for AI image generation in 2026. The suite offers four variants optimized for different use cases: FLUX.2 [max] for production-grade 4MP photorealistic output with multi-reference control, FLUX.2 [pro] for high-quality API use, FLUX.2 [flex] for flexible deployment, and FLUX.2 [klein] for sub-second inference on capable hardware. Unlike closed models, FLUX.2 offers open weights for self-hosting alongside a cloud API, making it uniquely versatile for developers and enterprises. The model excels at consistent character rendering, complex composition fidelity, and photographic realism — areas where it matches or exceeds Midjourney on benchmark comparisons. Access is available via the Black Forest Labs API with megapixel-based pricing (starting at $0.014/image for Klein, ~$0.03/image for Pro), through an open playground, or by self-hosting the open weights. The $300M Series B raised in December 2025 signals continued rapid development.
Fastest model, sub-second inference
$0.03/image at 1024x1024 (megapixel-based)
4MP photorealistic output with multi-reference
Explore other tools in this category
The leading AI image generator — now with V8, 5x faster and native 2K HD
OpenAI's advanced image generation model
Open-source AI image generation powerhouse
AI art generator with production-quality assets
Free AI image creator with intuitive canvas
AI image generator that nails text rendering — now with character consistency in v3.0
