Posts by Basem Barakat

DI Series: Serving Kimi-K3 MXFP4 Wide-EP Disaggregated on AMD Instinct MI300X and MI325X

• 08 October 2026
Ravi Gupta , Mir Ali , Doug Lehr , Basem Barakat , Markus Hartikainen , Nico Holmberg , Vikram Appia , Anuj Mittal , Onil Gunawardana , Aswin Mathews , Peng Sun , Emad Barsoum
English
Software tools & optimizations
AI/ML LLM Performance Serving Optimization Memory

This post is part of the AMD Disaggregated Inference (DI) Blogs Series, which takes frontier open models from a single-node demo to production, multi-node serving on AMD Instinct™ GPUs

Read more ...

---