Wen Xie#
Wen Xie is currently a PMTS Software Development Engineer in the Training at Scale team. His work focuses on developing and optimizing distributed training frameworks for large-scale models. He has more than 10 years of experience in performance optimization for both training and inference of image models and large language models, and is familiar with deep learning engineering optimization techniques.
Posts by Wen Xie
Enabling DeepSeek-V4-Flash Training on AMD Instinct MI355X GPUs with Primus
DeepSeek-V4-Flash training on AMD Instinct GPUs with Primus: model architecture introduction, performance projection, kernel optimizations, and how to reproduce.
Using ODC to Accelerate AMD SFT Training
Learn how we ported ODC on-demand P2P communication to AMD Instinct MI300X to cut FSDP bubbles and speed up variable-length SFT training.
Introducing Instella-MoE: A State-of-the-Art Fully Open Mixture-of-Experts Language Model
Explore Instella-MoE-16B-A3B, AMD’s fully open 16B MoE LLM with 2.8B active params per token, trained on AMD Instinct™ MI300 & MI325 GPUs.
Primus Tuning Agent: Closing the Configuration-Search Loop
Use the Primus Tuning Agent to automatically find optimal LLM training configurations on AMD Instinct GPUs.
Dropless MoE Training in JAX with Primus-Turbo
Learn how to train dropless MoE in JAX/MaxText with Primus-Turbo's grouped GEMM and DeepEP all-to-all for faster, more memory-efficient training.
Primus Projection: Estimate Memory and Performance Before You Train
Learn how to use the Primus projection tool to estimate memory and performance for large-scale LLM training on AMD Instinct™ accelerator platforms.
Primus-Pipeline: A More Flexible and Scalable Pipeline Parallelism Implementation
Learn how to use our flexible and scalable pipeline parallelism framework with Primus backend and AMD hardware.
Resilient Large-Scale Training: Integrating TorchFT with TorchTitan on AMD GPUs
Achieve resilient, checkpoint-less distributed training on AMD GPUs by integrating TorchFT with TorchTitan on Primus-SaFE.
Deep Dive into Primus: High-Performance Training for Large Language Models
Learn how to achieve peak dense LLM training performance on AMD Instinct™ GPUs using Primus’s unified CLI and optimized backend presets.
MoE Training Best Practices on AMD GPUs
Learn how to optimize Mixture-of-Experts (MoE) model training on AMD Instinct GPUs with ROCm. Maximize your AI training performance now!
Optimizing LLM Workloads: AMD Instinct MI355X GPUs Drive Competitive Performance
Explore ROCm 7.0’s AI training boost! See how MI355X accelerates JAX and PyTorch frameworks to unlock faster and efficient LLM scaling.
An Introduction to Primus-Turbo: A Library for Accelerating Transformer Models on AMD GPUs
Primus streamlines training on AMD ROCm, from fine-tuning to massive pretraining on MI300X GPUs—faster, safer, and easier to debug