Posts by Wen Xie
Using ODC to Accelerate AMD SFT Training
- 12 August 2026
Large-scale training spends a surprising share of its wall-clock time waiting instead of computing. Under Fully Sharded Data Parallel (FSDP), every layer ends in a collective all-gather or reduce-scatter, and every collective is a barrier that the whole data-parallel group has to reach together. Feed that machinery variable-length supervised fine-tuning (SFT) data and the picture gets worse: some ranks draw long documents while others draw short ones, so the fast ranks sit idle waiting for the slow ones.
Introducing Instella-MoE: A State-of-the-Art Fully Open Mixture-of-Experts Language Model
- 24 July 2026
AMD is excited to introduce Instella-MoE, a state-of-the-art fully open Mixture-of-Experts (MoE) language model with 16 billion total parameters and 2.8 billion active parameters. Trained from scratch on AMD Instinct™ MI300X and MI325X GPUs with the AMD ROCm™ software stack, Instella-MoE combines a sparsely activated MoE design with architectural innovations such as Gated Multi-head Latent Attention (Gated MLA) and FarSkip-Collective connectivity. Instella-MoE delivers competitive performance across a broad suite of benchmarks against both dense and MoE baselines (as shown in Figure 1) including models with comparable or larger active parameter counts, establishing it as one of the strongest fully open language models at its scale.
Primus Tuning Agent: Closing the Configuration-Search Loop
- 06 July 2026
Error parsing meta tag attribute “keywords”: No content.
Dropless MoE Training in JAX with Primus-Turbo
- 10 June 2026
Mixture-of-Experts (MoE) models have become a standard way to scale a transformer’s parameter count without paying the full compute bill — but training them efficiently on GPUs forces an uncomfortable trade-off. The default path in JAX/MaxText keeps every expert’s tensors at a fixed shape and simply drops the tokens that overflow each expert’s capacity, trading model quality for speed. The fully dropless alternative keeps every token, but in pure JAX it hits a memory wall that makes it impractical at production scale.
Primus Projection: Estimate Memory and Performance Before You Train
- 24 April 2026
Error parsing meta tag attribute “keywords”: No content.
Primus-Pipeline: A More Flexible and Scalable Pipeline Parallelism Implementation
- 23 February 2026
Error parsing meta tag attribute “keywords”: No content.
Resilient Large-Scale Training: Integrating TorchFT with TorchTitan on AMD GPUs
- 08 February 2026
Training large AI models on AMD GPUs demands unwavering stability and robust fault-tolerance capabilities at cluster scale. Yet today’s ROCm-based multi-node GPU deployments often rely on brittle checkpoint-and-restart mechanisms to recover from failures. This approach wastes precious compute cycles and slows down training as model sizes and cluster scales grow. To address these challenges, we integrated PyTorch’s native fault-tolerance framework—TorchFT—with the TorchTitan training framework on AMD’s Primus-SaFE Kubernetes platform, achieving resilient, checkpoint-less training at hundred-GPU scale. This blog builds upon our previous work on the Primus ecosystem—for background on the platform architecture, see our earlier posts on Primus-SaFE, the Primus training framework, and training large models with Primus.
Deep Dive into Primus: High-Performance Training for Large Language Models
- 15 January 2026
Primus is the AMD unified training framework designed to deliver high-performance, scalable large language models (LLMs) training across multiple backends – including TorchTitan and Megatron-LM. It provides a consistent CLI interface, while each backend ships with carefully optimized configurations for popular open-source models. These backend-specific presets ensure the best out-of-the-box performance on AMD Instinct™ GPUs. In this deep dive, we walk through the best practices for achieving peak performance when training dense LLMs on Primus.
MoE Training Best Practices on AMD GPUs
- 16 December 2025
This blog covers best practices for training Mixture-of-Experts (MoE) models on AMD Instinct™ MI300/MI355-series[a] GPUs with the ROCm ecosystem. Whether you’re new to MoE distributed architectures or optimizing trillion-parameter models, this guide will help you identify bottlenecks and maximize efficiency on AMD hardware.
Optimizing LLM Workloads: AMD Instinct MI355X GPUs Drive Competitive Performance
- 02 December 2025
AI training workloads are pushing the limits of modern GPU architectures. With the release of AMD ROCm™ 7.0 software, AMD is raising the bar for high-performance training by delivering optimized support for LLM workloads across the JAX and PyTorch frameworks. The latest v25.9 Training Dockers demonstrate exceptional scaling efficiency for both single-node and multi-node setups, empowering researchers and developers to push model sizes and complexity further than ever.
An Introduction to Primus-Turbo: A Library for Accelerating Transformer Models on AMD GPUs
- 19 September 2025
With the rapid growth of large-scale models, acceleration libraries are facing higher demands: they must deliver exceptional performance, offer comprehensive functionality, and remain easy to use. To meet these needs, we introduce Primus-Turbo — part of the Primus product family (see our previous blog for background). Primus-Turbo is designed around three core principles: performance, completeness, and ease of use. It supports training, inference, and a wide range of application scenarios, providing developers with a solid foundation to efficiently build and optimize large models on the ROCm platform. See Figure 1 below for a comprehensive stack coverage of Primus-Turbo.
Primus: A Lightweight, Unified Training Framework for Large Models on AMD GPUs
- 22 August 2025
Training large language models (LLMs) at scale is inherently complex. Different frameworks expose inconsistent interfaces, multi-GPU and distributed setups require brittle scripting, and backend-specific quirks introduce overhead that slows down training iterations. Primus tackles these challenges with a streamlined, backend-agnostic training framework that helps developers launch, customize, and scale training jobs faster on AMD GPUs.