在 NVIDIA Blackwell GPU 上加速 FLUX 3 Image Klein 图像生成
原标题：Speed Up Image Generation with FLUX 3 Image Klein on NVIDIA Blackwell GPUs
AI 摘要
NVIDIA 技术博客发布教程，介绍如何在 NVIDIA Blackwell GPU 上通过 TensorRT LLM VisualGen 加速 Black Forest Labs 的 FLUX 3 Image Klein 图像生成与编辑模型。该模型在 NVIDIA GB200 NVL72 上训练，结合 Qwen3-VL-4B-Instruct 文本编码、扩散 Transformer 和自编码器。在单张 B200 的 2048×2048 测试中，组合优化使 Base 模型的文生图提速最高 3.46 倍、图生图提速 4.45 倍，NVFP4 相比 BF16 也有约 1.6 倍提升。
正文节选
When deploying diffusion models in design tools, image editors, and visual production workflows, developers must account for inference latency. Generating an image requires multiple sequential passes through a model with billions of parameters, and higher resolutions increase attention costs. FLUX 3 Image Klein is an image editing and generation model from Black Forest Labs (BFL) trained on NVIDIA GB200 NVL72 and optimized for NVIDIA GPUs. It offers Base and distilled checkpoints for different l