Guide for Developer - intel/xFasterTransformer GitHub Wiki
Guide for Developer
GPU Part
- Enable pipeline parallel: https://github.com/intel/xFasterTransformer/pull/221
- Build w/ ICX: https://github.com/intel/xFasterTransformer/pull/228
- How to use GPU: https://github.com/intel/xFasterTransformer/pull/231
- Add GPU kernels and enable LLaMA model: https://github.com/intel/xFasterTransformer/pull/372
- Enable continuous batching on single GPU: https://github.com/intel/xFasterTransformer/pull/452