英伟达教程加速FLUX 3 Image Klein
NVIDIA Technical Blog 发布了在 NVIDIA Blackwell GPU 上通过 TensorRT LLM VisualGen 运行 Black Forest Labs 的 FLUX 3 Image Klein 的部署教程,涵盖 NVFP4、FP8 量化、融合内核、Cache-DiT、稀疏注意力和多卡并行。NVIDIA 称,在单张 B200 GPU 上进行 2048×2048 测试时,组合优化相较未编译的 BF16 推理最高可让文生图加速 3.46 倍、图生图加速 4.45 倍;这些性能数据属于厂商自测,并非独立基准结果。
教程提供了可用于 NVIDIA Blackwell 硬件的部署步骤和配置,可降低 FLUX 3 Image Klein 的推理延迟,但报告的加速效果依赖具体测试环境,并伴随可测量的输出质量权衡。
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 latency and fidelity requirements.
This tutorial shows how to configure FLUX 3 Image Klein in NVIDIA TensorRT LLM VisualGen to take advantage of fused kernels, multi-GPU parallelism, FP8/NVFP4 support, caching, and attention optimizations. It compares latency and changes in generated outputs across optimizations. In the reported 2048×2048 tests on a single NVIDIA B200 GPU, combined optimizations delivered up to 3.46x faster text-to-image (T2I) and 4.45x faster image-to-image (I2I) generation for FLUX 3 Image Klein Base compared with uncompiled BF16 inference. The tutorial includes deployment instructions and YAML configurations for applying these optimizations to your workloads.
FLUX 3 inference with TensorRT LLM VisualGen
FLUX 3 Image Klein combines Qwen3-VL-4B-Instruct for text conditioning, a diffusion transformer (DiT) to refine image latents, and an autoencoder to decode latents into images and encode I2I reference inputs. Klein processes text, noisy target-image, and optional reference-image streams independently before combining them through joint attention.
部分来源内容未展示,可前往原文继续阅读。
#Black Forest Labs#NVIDIA#FLUX 3 Image Klein#TensorRT LLM VisualGen#Qwen3-VL-4B-Instruct#NVIDIA Blackwell#NVIDIA B200#NVIDIA DGX B200#Cache-DiT#Sol-Attention