FLUX.2
A diffusion model series developed by Black Forest Labs. FLUX.2 klein is its 4-billion-parameter variant, designed for high-quality stylized image generation. The model can serve as a teacher model in knowledge distillation workflows.
Timeline (last 90 days)
The FLUX.2 series was launched by Black Forest Labs; the klein version has about 9B parameters and uses a rectified flow and MMDiT architecture.
The FLUX.2 [dev] version is a 32B parameter model capable of outputting 4-megapixel images; the [klein] 4B version can perform inference on consumer-grade hardware such as the NVIDIA RTX 3090.
The shift value of 2.8 follows the SD3/RAE rule calculated as √(32·32·32/4096), corresponding to the 32-channel latent space used by FLUX.2.
All Facts (3)
The FLUX.2 series was launched by Black Forest Labs; the klein version has about 9B parameters and uses a rectified flow and MMDiT architecture.
50%UnverifiedThe FLUX.2 [dev] version is a 32B parameter model capable of outputting 4-megapixel images; the [klein] 4B version can perform inference on consumer-grade hardware such as the NVIDIA RTX 3090.
50%UnverifiedThe shift value of 2.8 follows the SD3/RAE rule calculated as √(32·32·32/4096), corresponding to the 32-channel latent space used by FLUX.2.
50%