Stories by lmxyy

Radial Attention: O(nlogn) Attention for Long Video Generation with 2-4× Speedup

SVDQuant+NVFP4: 4× Smaller, 3× Faster FLUX with 16-bit Quality on Blackwell GPUs

RTX 5090 Workstation Configuration Journey

SVDQuant: 4-Bit Quantization Powers 12B Flux on a 16GB 4090 GPU with 3x Speedup