articleHuggingFace Blog
Creating custom kernels for the AMD MI300
Hugging Face and AMD optimize custom kernels for MI300X to boost Llama 3.1 405B inference in FP8 on 8 GPUs. They combine fused residual/RMS norm/FP8 conversion, SwiGLU, and a Skinny GEMM kernel, with benchmarks and open-source tooling in hf-rocm-kernels.
published JUL 09, 2025★★★★★
Read the sourcehuggingface.co/blog/mi300kernels
[*] Opens in a new tab · no tracking on Lantern's side
- Source
- HuggingFace Blog
- Ingested
- JUL 09, 2025 · 19:10
- Editorial score
- 4.0 / 5