view post Post 4299 gpt-oss was possible thanks to new engineering efforts in ๐ค transformers. We just dropped a blog covering them:- Kernels from the Hub- MXFP4 Quantization- Tensor & Expert Parallelism- Dynamic Sliding Window & Cache- Continuous Batching & Paged AttentionGrab a coffee & dive in! โ๏ธhttps://huggingface.co/blog/faster-transformers See translation ๐ฅ 12 12 ๐ง 2 2 ๐ 2 2 + Reply