FLUX.2 [dev] Turbo image model

FLUX.2 [dev] Turbo is a distilled LoRA adapter for FLUX.2 [dev] that enables high-quality image generation in just 8 inference steps. It was released by FAL.ai on December the 29th 2025.

Key Features

• 8-step inference - 6x faster than the base model's typical 50 steps
• Quality preserved - Matches or surpasses the original FLUX.2 [dev] quality
• LoRA adapter - Lightweight, easy to integrate with existing FLUX.2 workflows
• Multiple modes - Supports both text-to-image and image editing

At the time of writing it ranked #9 on the Artificial Analysis text-to-image arena just behind Seedream 4.5

The downsides might be the licence terms and high VRAM requirements.

The model can be accessed on FAL itself