You can use the docker image we provide ccr.ccs.tencentyun.com/tme-lyralab/lyradiff_workspace:1.0.0. docker run --gpus all -it --rm --name lyraDiff_workspace ccr.ccs ...
High performance: close to roofline fp16 TensorCore (NVIDIA GPU) / MatrixCore (AMD GPU) performance on major models, including ResNet, MaskRCNN, BERT, VisionTransformer, Stable Diffusion, etc. Unified ...
一些您可能无法访问的结果已被隐去。
显示无法访问的结果