Expert parallelism

5 posts · 3 sources

NVIDIABlog10 Jul

AI Model Co-Design: Hardware-Friendly LLM Design

BasetenBlog12 Nov

Kimi K2 Thinking at 140+ TPS on NVIDIA Blackwell

MetaBlog17 Oct

Scaling LLM Inference: Innovations in Tensor Parallelism, Context Parallelism, and Expert Parallelism

BasetenBlog29 Apr

Day zero benchmarks for Qwen 3 with SGLang on Baseten

BasetenBlog13 Feb

How multi-node inference works for massive LLMs like DeepSeek-R1